Skip to main content

Full text of "Study on Federal regulation: prepared pursuant to S. Res. 71,to authorize a study of the purpose and current effectiveness of certain Federal agencies"

See other formats


95th Congress 
2d Session 



L /v 



Prepared PuR&uii|T fro JAN 1 ^c 1378 / - < 

S. Res. ivkA J 






Volume VI 


Printed for the use of the Committee on Governmental Affairs 

^oj^e^^^®^^ ) COMMITTEE PBINT 

2d Session J 


Prepared Pursuant to 

S. Res. 71 





Volume VI 


Printed for the use of the Committee on Governmental Affairs 

83-944 WASHINGTON : 1978 


ABRAHAM RIBICOFF, Connecticut, Chairman 

HENRY M. JACKSON, Washington 
SAM NUNN, Georgia 
JIM SASSER, Tennessee 



WILLIAM V. ROTH, Jr., Delaware 


CHARLES McC. Mathias, Jr., Maryland 


H. JOHN HEINZ III, Pennsylvania 


Richard A. Wegman, Chief Counsel and Staff Director 

Paul Hoff, Counsel 

Theodore J. Jacobs, Counsel 

James M. Graham, Counsel 

Paul Rosenthal, Counsel 

Stuart Statler, Counsel 

Ethel Z. Geisinger, Special Assistant 

Harold C. Anderson, Staft Editor 

Regulatory Appointments 
James M. Graham 

Puilic Participation 

Mark Nadbl 
Ethel Z. Geisingeb 

Regulatory Organization 
Ethel Z. Geisixger 
James M. Graham 
Theodore J. Jacobs 

Congressional Oversight 
Paul Rosenthal 

Regulatory Delay 

William Pedersen 

Lawrence Novet 

Framework for Regulation 
Leonard Weiss 
Michael Klass 

Clerical Assistants 

Alberta Johnson 

Jean-Marie Peterson 

Advisory Panel for the Study 

Peter B. Hutt 

Harry C. McPherson 

Roger G. Noll 

Meeton J. Peck 

Robert Pitofsky 

William D. Ruckleshaus 

Lee C. Whitb 



U.S. Senate, 
Committee on Governmental Affairs, 

Washington, D.C. 
Hon. Walter F. Mondale, 
President of the Senate, 
U.S. Capitol, Washington, D.C. 

Dear ^Ir. President: It is a privilege to submit to you the appendix 
to the sixth and final volume of our study on Federal regulation. The 
appendix is composed of ten case studies, each of which considers Fed- 
eral regulatory activities in major areas of the national economy. The 
papers were prepared and written, at the request of the Committee, by 
leading experts in the field of Federal reguhition. The names of the 
authors along with brief biographical resumes are contained elsewhere 
in this volume. 

The Committee gratefully acknowledges the work of those individ- 
uals. We believe that their research and recommendations will make 
significant contributions to Congressional consideration of these im- 
portant issues in the future. The Committee also expresses its appre- 
ciation to Dr. Leonard Weiss and Dr. Michael Klass, who served as 
coordinators for this project. 

It merits emphasis, however, that the contents of the studies in this 
appendix are the sole responsibility of the authors, and that the facts, 
conclusions, and recommendations incorporated therein have neither 
been approved, disapproved, nor considered by the Committee on Gov- 
ernmental Affairs. 

Abe Ribicoff. 

Digitized by the Internet Archive 
in 2013 



Letter of transmittal iii 


Summary 1 

Abstract of regulation of overland movements of freight 3 

Federal regulatory history in the United States 3 

The act to regulate commerce (1887) 3 

The Transportation Act of 1920 8 

The Motor Carrier Act of 1935 11 

Current regulation 14 

What freight transport regulation is not 14 

What freight transport regulation is 25 

Impact of current regulation: costs and benefits 46 

Quantitative estimates of costs of regulation 46 

Qualitative estimates of the costs of regulation 53 

The current economic circumstances surrounding regulation 59 

Degree of modal transport monopoly 59 

Degree of competition within each mode 59 

Possibihties of predatory competition 60 

Transport contribution to economic development 60 

Transport contribution to competition or monopoly at other 

industrial stages 61 

Cross-subsidization 61 

Possible alternatives to the current regulatory scheme 62 

The railroads 62 

The motor carrier 64 

Conclusions 66 

Some conclusions about railroad regulation 66 

Some conclusions about motor carrier regulation 68 

Suggestions for reform of motor carrier regulation 70 


Summary 75 

Domestic trunk airline regulation: An economic evaluation 77 

Scope of the present study 77 

The airline industry 79 

The development of airline regulation — A brief historical sketch 80 

CAB pohcies in administering the Civil Aeronautics Act of 1938 87 

Policies toward new firm entry 87 

Regulation of fares 88 

Service quality competition 93 

The domestic passenger fare investigation — scope and intention. _ 94 

The impact of CAB regulation: An economic analysis 95 

Evidence of the impact of CAB regulation on fares and service 

qualities — the intrastate routes 95 

Cost-based trunk fares versus actual intrastate fares 99 

Optimal faros and product quality: The contributions of Douglas 

and Miller 103 

The existence and optimahty of competitive markets in air 

transportation 107 

The economic effects of CAB regulation on consumers 117 

The impact of CAB regulation on producers 123 

The impact of regulation on managerial efficiency 125 

Regulation and technological change . 127 



Domestic Trunk Airline Regulation: An Economic Evaluation — Continued ^^B9 

Goals and achievements of trunk airline regulation — an evaluation _> 128 

Economic efficiency 128 

Cross-subsidization 128 

Needs of post office and national defense 131 

Economic conditions of the carriers 132 

Destructive competition 133 

Safety 133 

Policy alternatives 134 

Improved regulation 134 

Deregulation by the CAB within the existing legal framework__ 135 

Nationalization 137 

Complete and immediate deregulation 137 

Partial and/or gradual deregulation via legislative reform 138 

Concluding comments 149 

Appendix: Calculations of welfare effects of CAB regulation 150 

Fare regulation, ffight frequency, and service quality 150 

Cost estimates 151 

The fare costs of regulation 155 

Impact of regulatory policies on service quality 157 

First-class losses 159 




Executive summary 163 

The Occupational Safety and Health Administration — an overview 169 

Introduction 169 

Statement of the problem 169 

Outline of the analysis 169 

Qualifications regarding this study 170 

Occupational safety and health: Market performance and the Gov- 
ernment's role 171 

Safety and health determination in perfectly competitive markets. 171 

Appropriate levels of safety and health 174 

Market imperfections 178 

Regulatory intervention 182 

Government involvement in occupational safety and health 191 

Involvement prior to OSHA 191 

The Occupational Safety and Health Act of 1970 194 

OSHA's performance 198 

Implementation 198 

Impacts 211 

Recommended directions for reform 222 

OSHA's role 223 

Reform of OSHA 225 

Concluding remarks 235 

Postscript 237 

Introduction 237 

Ehminating the nuisance standards 238 

Increased emphasis on health. 239 

The interagency task force 243 

Concluding remarks » 244 

Bibliography 245 

Individuals interviewed for this report 248 


Introduction and summary 251 

Federal environmental regulation 255 

The Environment, the regulatory process and the market 255 

Environmental versus "economic" regulation 255 

Market forces in environmental poUcy 262 

Evolution of U.S. environmental regulation 272 

Pre-1970 Federal legislation 272 

Environmental legislation of the early 1970's 275 

General observations on the 1970's policy directions 282 


Federal Environmental Regulation — Continued I*aff« 

Evaluation of the 1970 Clean Air Act 283 

Implementation of the 1970 Clean Air Act 283 

Air pollution: Trends and prognosis 294 

Summarv evaluation of the 1970 CAA 298 

Evaluation of the 1972 Federal Water Pollution Control Act 299 

Implementation of the 1972 Fedeial Water Pollution Control Act. 299 

Water pollution: Trends and prognosis 306 

Summary evaluation of the 1972 FWPCA 308 

The lessons of recent experience 309 

Ambient indices, the "threshold theory" and environmental 

quality 309 

Regulating technology over time 314 

The costs and benefits of environmental protection 321 

ImpUcations for regulatory revision 327 

Market systems in environmental pohcy 330 

Pohtical objections to market systems 330 

Managing air quahty 333 

Managing water quality 338 

ControlHng toxic pollutant? 340 

Environmental pohcy in the long run 344 

CABLE television: the framework of regulation 


Summary 347 

Cable television: The framework of regulation 349 

Introduction 349 

Historical background 353 

Forces and issues 357 

The current regulation scheme 360 

Outhne of FCC cable rules 361 

The impact of cable regulation 365 

Economic effects 366 

Pohcy studies 373 

Alternatives 375 

Conclusion 378 

Appendix 38 1 

General conclusions 381 

Franchising considerations 381 

Pay television 381 

Cable access to programing 382 

Ownership 382 

Federal regulation 382 

State regulation 383 

General considerations 383 

Long-range polic}^ recommendations : 

Industry structure: Distribution 383 

Industry structure: Programing 384 

Institutional and jurisdictional framework for cable regulation. _ 384 

The consumer and the cable 384 

Cable ownership and access 385 

Copyright 385 

Pay cable 385 

Two-tier regulation 385 

Basic principles 385 

Specific recommendations to carry out those principles 387 

BibHography 388 


Introduction 393 

Regulation of the U.S. food industry 397 

History of Federal regulation of the food industry 397 

The evolution of Federal laws 397 

The concept of adulteration: Health versus aesthetics 399 

The purposes and basic mechanisms of consumer protection regulation. 402 


Regulation of the U.S. Food Industry — Continued : Page 

Regulatory mechanism of FDA 421 

Introduction 421 

Current regulatory scheme 422 

Role of the judiciary 424 

Consequences of concentration on good manufacturing practices- 424 

The question on sanitation once again 427 

A model of the effects of FDA inspection 429 

Effects of FDA regulation: Theory 431 

Evidence on the effects of FDA 435 

Regulatory mechanism of the Federal Trade Commission 445 

FTC enforcement procedures 445 

FTC and the food industry 446 

FTC and nutrition 450 

The role of the USDA 463 

Meat and poultry inspection 463 

Overview of plant inspection 464 

Analysis of inspection procedures 465 

Information approaches 468 

Overlap with the FDA 471 

Conclusions 471 

Conflicts between Federal, State, and local food regulations 472 

Summary . 478 

References :. 482 



Summary 487 

Glossary ^ 491 

An evaluation of Federal milk price regulation: History, impact, and 

options for reform 495 

I. The Preregulation 1930's dairy industrj^ 495 

Milk production on farms 496 

Milk processing and distribution 496 

Market structure and the role of cooperatives 498 

Market stabiUty: Annual fluctuations in production and 

prices before the depression 500 

Impact of the depression on milk markets 502 

Summary: The preregulatory dairy industry 504 

II. The origins of Federal milk marketing regulation: The 1933 
Agricultural Adjustment Act and the 1937 Agricultural 

Marketing Agreement Act 505 

III. Basic provisions of Federal milk marketing orders 506 

Pricing under Federal orders 507 

Dairy cooperatives under the Federal order system 516 

IV. Supply control 519 

Supply control at the national level 519 

Supply control at the regional level 523 

Supply restrictions at the local level 529 

Summary of effects of national, regional and local restric- 
tions 531 

V. The 1978 case for reform of Federal milk marketing order and 

policy toward cooperatives 532 

The theory: How classified pricing and supply control alter 

milk production and consumption 532 

Empirical studies and estimates of the effects of Federal 

orders and cooperatives 536 

Indirect evidence of the impact of Federal milk marketing 

orders and cooperatives 542 

Summary of the case for reform of Federal milk marketing 

orders and policy toward cooperatives 1 545 

VI. 1978 milk industry: How much has changed since the 1930's? 545 

Local milksheds turning into one national market 549 

Market power of dairy farmers versus processors and/or ^^ \ 

distributors 549 

Orderly marketing conditions and stability 551 

Effects of orders on the availability of cheap, high-quality 

milk to consumers 555 

Conclusion: How much has changed since the 1930's? 556 


All Evaluation of Federal Milk Price Regulation: History, Impact, and 
Options for Reform — Continued 

VII. Options for reforming Federal milk marketing order price regula- Paw 

tions and curbing cooperative market power 556 

The criteria 557 

Transitional disruptions caused by changes in the 

regulations 558 

Overall recommendations 566 

Government roles in the milk industry in the absence of 

Federal orders 567 

Appen dixes 570 

Bibhography 579 



Summary 589 

Regulation of securities markets: An examination of the effects of increased 

competition 593 

Background 593 

Functions of securities markets and criteria of efficiency 593 

Markets today 594 

Historic background: The New York Stock Exchange 595 

Historical background: Regulation 596 

Cartel pricing of brokerage services prior to May 1, 1975 599 

Characteristics of the NYSE cartel 599 

Effects of fixed commission rates and arguments in favor of com- 
petitive rates 607 

Arguments against competitive commission rates 615 

Economics effects of competitive commission rates 622 

Commission rates 623 

Broker revenue? and profits 629 

Structure of the industry and destructive competition 633 

Market fragmentation 641 

Research and unbundling 642 

Liquidity and market quality 644 

Related developments 645 

Appendixes 647 

References 655 


Summary 659 

An Examination of regulation in the natural gas industry 663 

Introduction 663 

Effects of policy changes: The time lag 666 

Historical background 666 

Conditions leading to the Natural Gas Act 667 

Regulation reaches the wellhead 670 

The current regulator}' scheme 673 

Economic forces in wellhead markets 676 

Reserve accumulation and extraction 677 

The demand for gas 679 

The forces and the shortage 679 

Impact of regulation 681 

The current gas shortage 681 

Some causes of the shortage 687 

Wellhead price regulation: Some qualitative considerations 690 

Wellhead price regulation: Some quantitative effects 695 

The nature and results of simulations 700 

A critique 702 

Reexamining the need for regulation 707 

Local distribution and policies 707 

Structure of wellhead markets 709 

Alternatives for social control of natural gas 714 

Some alternative measures 715 

Selected references 726 




Summary 731 

A case study of regulatory programs of the Federal Energy Administration. 733 

The situation from August 1971 to August 1973 733 

Price controls — January to August 1973 741 

August 1973 to December 1975 743 

The Energy Policy and Conservation Act 746 

Regulations under the Emergency Petroleum Allocation Act 749 

Crude oil price controls 750 

Crude oil entitlements 753 

Refined product price controls 759 

Crude oil allocation 761 

Refined product allocation 763 

Evaluation of 1974-75 controls 765 

Crude oil price controls 766 

Product price controls 786 

Summary: The EPAA regulatory program 796 

Current oil price controls 799 

Crude oil price controls 799 

Entitlements 805 

Evaluation of the price control program 812 

Crude oil price controls 812 

Product price controls and mandatory allocation 817 

Purchaser-supplier freeze 820 

Small refiner bias . 820 

Costs of paperwork and complexity 821 

Exceptions from regulations 822 

Effects of permanent controls 824 

Summary : Current regulatory programs 824 

An alternate view of price controls 827 

Alternatives to oil price controls 828 

Maintain crude oil price controls 829 

Accelerated removal of price controls 830 

Excise taxes on crude oil 831 

Appendix 833 

Professional background of authors 835 


(By James R. Nelson, Amherst College) 


The main themes of this discussion are derived from history, from 
economic analysis, and from the institutional features of the everyday 
world of transporting goods by rail or road. 

The historical discussion begins by emphasizing the extent to vrhich 
original regulatoiy policy in the United States was GP<;ablished by the 
chaTracteristics of the railroad industry and of public opinion with 
respect to that industry in the ISSOs. It continues by examining the 
extent to which the major regulatory changes introduced in 1920 were 
still related almost solely to the idea of ''railroad monopol3\'- It con- 
cludes by pointing out the extent to which the Motor Carrier Act of 
1935 was designed to protect the railroads, and based on the irrelevant 
assumption that, since trucks compete with railroads, they must be 
regulated to the greatest extent possible as if they were railroads. The 
conclusion of the historical section is that regulation of railroads had 
ample economic as well as political justification when it was fii^t at- 
tempted, but that neither the law nor the Interstate Commerce Com- 
mission lias really come to terms with the internal combustion engine. 

The next section, which is devoted to the characteristics of current 
regulation, begins with a discu^jsion of what current transport regula- 
tion is not. Unlike public utility regulation, it is not regulation of rates 
of return. In the railroad case, this is true in large part because of 
doubts on all sides as to whether anything close to a "normal'' return 
is obtainable, and in the motor carrier case because an industry con- 
taining so many firms, and so m.uch route duplication, is not an in- 
dustry in which administrative standards for rate of return can be 
applied with any degree of nicety. In spite of Interstate Commerce 
Commission control over both financial investment, in the form of 
new cnpital issues, and real disinvestment, in the form of railroad 
branchline abandonments, regulation also has only tangential in- 
fluence on the amount and direction of new investment. Further, on 
the I.C.C.'s own showing, regulatory control over rates is practically 
effective with respect to a very small fraction (less than one percent) 
of proposed rate changes. When attention is shifted from the depth 
to the breadth of control, it becomes evident that traffic has steadily 
moved from the only transport mode which is 100 percent common 
carrier and 100 percent regulated (the railroads) to unregulated truck- 
ing, or to barge lines and pipelines which are subjected to very little 
control of the thoroughgoing form applied at least since 1920 to 

This section could easily lead to the conclusion that regulation is 
neither good nor bad, but simply superfluous. However, this would 
need qualification. For one thing, the Interstate Commerce Conunis- 
sion claims to have saved the public very large sums of money by 
reducing railroad requests for rate increases. And with respect to 
rates, the Interstate Commerce Commission's casting itself in the role 
of the innocent bystander is overdrawn in several respects. The num- 
ber of rates on wliich the Commission takes affirmative action is ob- 



viously a minimal test of the cumulative importance of these rates. 
Moreover, the Interstate Commerce Commission is an integral part 
of the apparatus by means of which new rate proposals by individual 
carriers are finally transmuted into effective rates. The Interstate 
Commerce Commission also has important control over institutional 
structure, in the motor carrier industry, due to its controls over entry. 
Its control over service standards has been concentrated on piecemeal 
rearguard actions with respect to railroad abandonment. Interstate 
Commerce Commission controls over freight cars have fallen well 
short of success for reasons largely beyond regulatory control, notably 
those stemming from managerial fragmentation of the American rail- 
road industry. 

This is not a very inspiring picture. But the economics profession 
has probably managed to caricature it. Specifically: (1) Economists 
tend to exaggerate the importance, for good or ill, of transport regu- 
lation. (2) Economists tend to quantify the qualitative. Regulation 
does have costs, and regulation does have benefits ; but neither of these 
can be measured by the pound. (3) Economists tend to cater to the 
greatest weakness of regulation: the attempt to penetrate the future 
by burrowing deeper into the past. Published estimates of the "costs of 
regulation" (or the "benefits" thereof) should be heavily discounted, 
if not disregarded entirely. 

Once one peers beneath the numbers to find out what is really hap- 
pening, the basic defect of regulation would appear to be not its im- 
position of huge extra costs but its failure to contribute to identifiable 

^^^at has this to suggest about regulatory policy? 

As to railroads, the basic issues have already been recognized in 
two ways. The first has been the appearance of government financial 
assistance, as exemplified by Amtrak, by government participation 
in the establishment of ConRail, and by other provisions for govern- 
mental financial assistance. The second has been the passage of new 
regulatory legislation which has culminated in the so-called "4-R" 
Act of 1976. This Act will increase the regulatory burden for a long 
time to come. But it does represent a modernized approach to rail- 
road regulation in so far as it pays more attention to the need for 
railroad profitability, the need for government assistance if this prof- 
itability cannot be attained on what are deemed to be important public 
services, and the need for more managerial discretion. 

As to motor carriers, the situation still appears to be one of stale- 
mate, although this stalemate may be broken by changes now before 
congressional committees. Part of this stalemate may reflect the prac- 
tical application of an economist's theoretical notion : if the economic 
and technical characteristics of an industry are such that competition 
is bound to work, then, even in the face of regulation, many aspects of 
competition probably will work. The trouble is that the part without 
the whole may produce perverse results. Re,gulation of motor carriage 
was first introduced to protect the railroads. Its present contribution 
to this objective is probably minimal. Meanwhile, it is in fact protect- 
ing all kinds of other people. A g:eneral loosening of motor carrier reg- 
ulation would not only yield direct economic benefits; it would also 
help to reveal who these protected interests are, what they are doing, 
and what they might be doing with less or no regulation. 


(By Ja^ies R. Xelsox, Amherst College) 

The major purposes of this paper are as follows : 

(1) To describe the main events in the history of Federal regulation 
of rail and motor carriers of freight. 

(2) To relate the main stages of this historical development to the 
economic environment in which they occurred, and to examine the 
question of the extent to which an economic institution, like any other 
facet of the economic system, may become functionally obsolete while 
retaining its technical efficiency. 

(3) To prepare the way for a discussion of regulation in its present 
economic environment by sorting out as carefully as possible the costs 
and benefits of present regulatory laws and practices, and of any feas- 
ible regulatory laws and practices. 

(4) To present alternative suggestions as to directions in which 
regulation might move — when, in what order, and how nnich. 

Federal Regulatory History in the United States 
the act to regulate commerce (1887) 

The growth of the American railroad system, coincident with the 
abrupt check to more extreme states' rights positions which occurred 
in 1865, produced strong Congressional concern with respect to rail- 
road rates, earnings, investment, and service. The first result was 
the Windom Report, "Report of the Select Committee on Transporta- 
tion — Routes to the Seaboard" (43d Congress, 1st session. Senate Re- 
port 307; Washington, G.P.O., 1874). The Select Committee devoted 
thirteen pages (pp. 109-122) to the consideration of "Competition 
between Railways, and Its Promotion by the Construction of Addi- 
tional Lines," and eighteen more pages (pp. 122-140) to "Direct Reg- 
ulation by Congress," before opting for its favorite : "Indirect Regu- 
lation and Reduction of Charges, through the Agency of One or 
More Railwav Lines To Be Owned, or Controlled, by the Government" 
(/6/fZ., pp. 140-161). 

The Cullom Committee Report,^ the immediate precursor of the 
Act to Regulate Commerce, appeared only a dozen years after the 
Windom Report. But it seems to belong in a different world. Com- 
mission regulation was to be found throughout the country; but it 
was concentrated, significantly, in a Granger belt consisting of Illi- 
nois, Wisconsin, Minnesota, Dakota Territory, Nebraska, Kansas, 
Missouri, and lowa.^ 

1 Report of the Senate Select Committee on Interstate Commerce, 49th Congress, Ist 
sess.. Senate Report 46, Washington, D.C., G.P.O., 1886. 
'Ibid., p. 65. 


The Cullom Committee sums up "The Causes of Complaint against 
the Kailroad System" under eighteen headings,^ and then condudes : 

. . . the most important, and in fact nearly all, of the foregoing complaints 
are based upon the practice of discrimination in one form or another.* 

A further theme, stressed by railroad ofiicials but embraced by other 
witnesses with "unanimity," was "construction of unnecessary rail- 
roads for speculative purposes." ^ Part of the light-density branch 
line problem was already there on the day the line first opened. 

The kind of competition favored by the witnesses before the Cullom 
Committee — including most of the witnesses who represented state 
railroad commissions — was "yes, but" competition, or competition 
without complete freedom to discriminate. The paradox in this posi- 
tion lay in the fact that the essence of railroad competition in the 
1880s was its differential character, and the essential product of 
differential com^petition is discriminatory pricing. 

If Senator Cullom lent his name to the most famous of Congres- 
sional reports on the subject of railroad regulation, Representative 
John Reagan of Texas was at least as influential in helping to deter- i 
mine the final form of the Act to Regulate Commerce. In May 1878, 
without much experience in rail or regulatory matters, he introduced 
his first bill, providing that : 

. . . freight rates and facilities would be made equal for all shippers, rebates 
and drawbacks would be prohibited, '"pools" or combinations formed to elimi- 
nate competition and to pool earnings would be forbidden, and charges for a 
short haul of goods (if as large as a carload) would not be greater than for a 
long haul . . . railroads would have to post freight rates and schedules. . . ." 

The first idea is ambiguous, but it can nevertheless be traced through 
into Sections 1, 2, and 3 of the Act to Regulate Commerce. The other 
ideas have become the common currency of railroad regulation — in 
qualification, as well as in application — since 1887. And, on a very 
critical issue which arose eight years later : 

The Reagan Bill explicitly forbade pooling ; the Cullom Bill said nothing on the 
matter. Reagan, when in conference with the Senators to reconcile the two bills 
in December 1886, conceded on virtually every point of disagreement save 
pooling. . . .'' 

For tliis and other reasons (e.g.. no control over entry into the indus- 
try, no powers of rate setting for the Commission) , one must agree Vv'ith 
tile Hilton verdict that, "as an effort at stabilization of the railroad 
cartels of the late nineteenth century," the Act to Regulate Commerce 
was "grossly inadequate." ^ But one need not agree with either pro- 
competitive economists, like George Hilton, or Marxist historians, like 
Gabriel Kolko, that the only or even main point to the Act to Regulate 
Commerce was to stabilize cartels. 

Many of the most important sections of the original Act to Regulate 
Commerce vrere altered beyond recognition, or destroyed, by subse- 

3n)id., pp. 180-181. 

4 Ibid., p. 182. 

«Tbid.. p. 48. 

« Ben H. Procter, "Not Without Honor : The Life of John H. Reagan," University of 
Texas Press, Austin, Tex., 1962. p. 226. 

■^ Gabriel Kolko, "Railroads and Regulation : 1887-1916." Princeton University Press, 
Princeton, N.Y., 1965, p. 43. 

"Pooling" could take one or both of two forms : division of traffic in agreed proportions ; 
and divisions of gross or net revenues in agreed proportions, independently of the original 
flow of trafHc. 

8 George W. Hilton, "The Transportation Act of 1938 : A Decade of Experience," Indiana 
University Press, Bloomington, Ind., 1969, p. 4. 

quent Supreme Court decisions and had to be restored in their original 
or in altered form by later legislation. But the fxrst four provisions 
survived, in principle, and are still very much in existence in practice. 
Section 1 generalized by requiring all rates to be "just and reasonable*' 
and declaring that "every unjust and unreasonable charge" is unlaw- 
ful. Section 2 prohibited personal discrimination "in the transportation 
of passengers or property" (which did not, of course, discourage elab- 
oiate systems of classification of commodities and passengers). Section 
3 prohil)ited ". . . undue or unreasonable preference or advantage . . . 
or any undue or unreasonable prejudice or disadvantage. . . ." And 
Section 4, the Long-and-Short Haul Clause, attempted to be m.ore 
specific about one supposed form of this discrimination by (as its pop- 
ular description indicates) forbidding higher rates for shorter than 
for longer hauls "under substantially similar circumstances and condi- 
tions." Finally, Section 5, which was in effect repealed in 1920, pro- 
hil)its pooling. 

What, then, were the main features of the ecor.omic landscape when 
the Act to Eegulate Commerce was passed, in 1S8T, which have since 
disappeared ? And what principal features have remained ? 

Economic ohsolescence : Railroads then and now 

(1) It was not universally true in 1887 that "the railroads had 
a monopoly." They faced intense competition from water carriers on 
some nv^u^s and on the Great Lakes and along the coasts. At the time 
the Cullom Committee submitted its report, even the Erie Canal was 
sufficiently important to influence railroad rates on grain as well as 
other bulk connnodities. Both the Windom and the Cullom Committees 
placed a stress on waterways which would have pleased the devotees 
of internal improvements in the 1830s. A perplexing question for the 
late nineteenth century was : how can one reconcile maximizing water 
competition and minimizing railroad rate discrimination ? One could 
not then, and one cannot now. Both the geography and the economics 
of water carriage practically force railroads to discriminate if they 
are to compete. 

(2) It was true in 1887 that the railroads had a unique position as 
common carriers which cannot be duplicated today by any one mode 
of transportation — or perhaps even by all modes taken together. First, 
as to passenger service: railroad passenger revenue in 1890 was $261 
million, as compared with $714 million for freight traffic.^ Thus pas- 
senger service created over one-quarter of all railroad revenues. Rail- 
roads not only dominated passenger service; they also dominated the 
movement of mail, express, and baggage. So, with respect to move- 
ments other than freight, the history of railroad regulation can be 
traced back not only to the cost characteristics of railroads themselves, 
but to doctrines emanating from stagecoach and turnpike days when 
rates vs'cre separated from tolls, and passenger service standards were 
of great relative importance. 

(3) Railroads not only had a great deal to do with the location of 
production, and to a lesser degree of consumption, but also with the 
location of various intermediate functions. ^Vholesalcrs were some of 
the most eloquent witnesses before the Cullom Committee. And rail- 

"U.S. Department of Commerce, Bureau of the Census. "Historical Statistics of the 
United States," Bicentennial Edition, G.P.O., Washington, D.C., 1975, Part 2, pp. 730 and 

83-944—78 2 


road rate structures could determine the fate not only of wholesalers 
in the strict sense, but also of wholesale buyers.^° Kelative rates and 
relative service mattered. And they mattered not only by location and 
by commodity, but also by the extent (if any) to which carload rates 
were below those on less-than-carload shipments. 

(4) Kailroads were expected to be able to influence the degree of 
competition at other levels of economic activity. Large individual 
shippers had a bargaining edge; large shippers who were expanding, 
and hence presumably always in search of new plant sites, had a fur- 
ther bargaining edge. Farmers, scattered across the county side, 
clearly had no such advantai^es; nor did country elevators. Therefore 
the popular feeling against the railroads themselves for price discrim- 
ination was exacerbated by feeling against what the railroads might do 
to competition at other levels of industry by acceding (even against 
their will) to monopsonistic pressures. This was no mere generaliza- 
tion to Congressmen who served between 1870 and 1887; they always 
had before them striking examples such as that of the Standard Oil 

^' . -, ^ 

(5) Many students of "the railroad problem" were not only alarmed 

at the discriminatory aspects of railroad competition, but also by 
its possibly self-defeating aspects. They lived in a world of both 
parallel and especially end-to-end railroad mergers. If no transcon- 
tinental system resulted, it was not for want of trying on the part of 
such railroad kings as Collis P. Huntington and (later) George 
Gould and E. H. Harriman. 

(6) If public opinion was opposed to over-building in general, it 
had no such opposition to new railroad construction which connected 
specific towns for the first time to the national network. Moreover, 
every town wanted as many lines, preferably under as many different 
ownerships, as possible. So entry control could not possibly have been 
included in the Act to Regulate Commerce in 1887. 

(7) Behind all these points lay the basic facts about railroading as 
an industry. It was intensely dynamic, in every possible sense of the 
word. It was expanding rapidly if irregularly in mileage, in number 
of trains operated, in volume of both passengers and freight, in reve- 
nues, in technical accomplishments — but not necessarily in financial 
strength. For one thing, the railroads' period of most rapid relative 
expansion between 1865 and 1895 occurred during a period of price 
recession in the United States which was then unmatched and may 
never again be matched anywhere in the world. For another, dynamic 
pricing might produce results which would make no sense for a mature 
industry. With a rapidly-expanding market and with unit costs which 
decreased with volume, decreased with improved technology, and de- 
creased along with the general price level, there could be an economic 
case for pricing below current costs. What may be the definitive study 
of railroad competition in Official Territory before 1900, Paul W. 
MacAvoy's "The Economic Effects of Regulation," " brings out the 
impetus toward intensive railroad rate competition generated by (1) 
the prospect of diverting very large tonnages from the Great Lakes ; 

10 See the thorough discussion of both of these points in George H. Miller, "Railroads 
and the Granger Laws," University of Wisconsin Press, Madison, Wis., 1971 — especially 
Chapter Five (pp. 97-116). "Iowa : The Revolt of the River Towns." 

11 Subtitled "The Trunk-Llne Railroad Cartels and the Interstate Commerce Commission 
Before 1900." The M.I.T. Press, Cambridge, Mass., 1965. 

(2) new entrants; (3) differential competition on the through 
routes — New York, Philadelphia, and Baltimore to Chicago— as com- 
pared with local routes. In a rapidly-growing industry which is at 
the core of a rapidly-growing economy, current costs and revenues 
may be largely irrelevant when compared with future prospects. 

Economic persistence : Railroads then and now 

One main point to this paper may be expressed in the brief state- 
ment that railroad similarities, from 1887 to the present, are consid- 
erably less marked than railroad differences. 

But a few underlying characteristics have remained : 

(1) The intense concern with railroad rate discrimination which 
was a national attitude in the 1880s is with us in diminished form 
toda3\ Wholesalers do not care as much ; commercial centers have lost 
some of their interest; but port authorities care, as do shippers and 
consignees of bulk commodities, unless they happen to depend entirely 
on waterways. However, waterway improvmonts greatly altered the 
1887 railroad discrimination problem, in part by enhancing it in some 
sections of the country; motor carrier competition has knocked the 
top off the discrimination problem, by taking over the highest-rated 
commodities and by substituting ubiquity for what used to be ample 
possibilities for geographical favoritism. Eailroad rate discrimination 
is still with us, and still a matter of concern. But it is no longer capable 
of restructuring the industrial map of the United States. 

(2) The combination of objection to over-building in general with 
enthusiasm for over-buildinor in particular still remains. But in the 
1880s the idea of over-building was often associated with new con- 
struction. It now relates to retaining trackage already in existence. 

(3) The railroads were Balkanized in 1887. In spite of many merg- 
ers since — and, in part, because of an increase in long-distance rail 
movements since— they are still Balknnized today. In the 1880s, how- 
ever. Balkanization was viewed as a protection for the public against 
Universal Bailroad ^Monopoly. This danger is no longer regarded as 
so acute — if acute at all. 

(4) Xo one showed much analytical interest in the relationship 
between railroad rates of return and capital flows in 1887, and this 
is still not a prime object of interest today. This was readily compre- 
hensible in 1887, since "watered stock'' was a major railroad contri- 
bution to corporate finance and for this and other reasons reported 
railroad rates of return were not to be trusted. But, on the assump- 
tion that private enterprise is expected to carrv on railroad service, 
disinterest in present low profits is incomprehensible today. 

The events leading up to the Act to Regulate Commerce have been 
discussed at some length because in railroading, more than in any 
other American industry, heredity matters. Given the essentially 
legalistic character of the regulatory process and the necessary reli- 
^f'^^r/r 1 r^¥^^^ ^^ ^^^e ^^^^^ ^n what has been decided in the past, 
the dimculty is not so much that regulation reaches incorrect solu- 
tions. It is, rather, that regulation is constrained both by history and 
by its own nature from planning ahead to avert future problems, or 
even addressing itself promptlv to immediate problems. 

Legislation with respect to railroad regulation from 1903 through 
1913 was mainly designed to patch rents in the fabric of the Inter- 
state Commerce Act which had been created by Supreme Court deci- 


sions. The El kins Act of 1903 Avas probably an exception : it tightened 
the requirements applyino- to personal discrimination (Section 1) = 
But the Hepburn Act, of 1906, superseded Supreme Court decisions 
by restoring to the Interstate Commerce Commission the power to 
establish -maximum railrond rates which the Commission had thought 
it had from the outset. The Mann-Elkins Act of 1910 also repaired 
a large hole in Section 4 ("Long-and-short Haul") which had re- 
sulted from a Supreme Court decision, and gave the Interstate Com- 
merce Commission the very important power to suspend a proposed 
rate before it came into effect. 


The Transportation Act of 1920 introduces the Modern Era Avith 
respect to all of these points. 

First, the Act of 1920 was a retrospective piece of legislation. It 
attempted to encompass all of the regulatory and economic changcB 
that had occurred since 1887 — and to relate them to what was still 
believed to be essentially a monopoly railroad industry. 

Second, in terms of what it set out to do, the Act of 1920 was the 
most complete and systematic transport regulatory measure ever 
passed by Congress. For the first and last time, the critical questions 
were posed: Assuming that one really vv'ants to regulate railroads (as 
many states already regulated public utilities), what does one do? 
Specifically, what problems does rail transportation present? And 
how may they be resolved ? 

What was the attempted contribution of the Transportation Act 
of 1920 ? And what Avas its actual contribution ? 

The purported keystone of the Act is well summiarized in the title 
of the book by William JST. Leonard : "Railroad Consolidation under 
the Transportation Act of 1920." ^" The desideratum was not One Big 
Railroad; nor was it — as in Britain and France at approximately 
the same time — a feAv large regional railroads with minimum com- 
petitive overlap. The goal was typically American: to equalize com- 
petition by incorporating the weak into the strong. It was still appre- 
ciated that this competition would be difierential competition, with 
more benefits in rates and services at points of convergence of sys- 
tems than at points of divergence. The only new constraint on the 
possible effects of this differential provision was to grant the Inter- 
state Commerce Commission ncAV powers to establish minimum rates. 

Pending the achievement of a consolidation plan which would equal- 
ize competitive conditions, the Act of 1920 both stated a final goal 
and advanced certain ameliorative proposals which were expected to 
be of use both in preparing for consolidation and in advancing the 
final goal. 

The final goal was expressed in Section 15a : 

In the exercise of its power to prescribe just and reasonable rates the Com- 
mission shall initiate, modify, establish or adjust such rates so that carriers 
as a whole (or as a whole in each of such rate groups or territories as the Com- 
mission may from time to time designate) will, under honest, efficient and 
economical management and reasonable expenditures for maintenance of way, 
structures and equipment, earn an aggregate annual net railway operating income 
equal, as nearly as may be, to a fair return upon the aggregate value of the- 

"Columl)la University Press, N.Y.. 1946. 

railwr.y property of such carriers held for and used in the service of 
transportation. ... 

Therefore the Act had two mutually supportive principal objectives : 
(1) to enable carriers "as a whole" to earn "a fair return"; (2) to 
enable the Interstate Commerce Commission to reorganize the railroad 
industry in such fashion that a fair return, overall, would also mean 
a fair return for each component. 

Certain other provisions of the Act of 1920 were designed both as 
temporary measures pending the full applicability of the '^Rule of 
Rate Making" and the provisions for railroad consolidation, and as 
permanently important in their own right. These included a direction 
to the Commission, in its division of joint rates, to consider "the 
Amount of revenue reouirod to pay their respective operating expenses, 
taxes, and a fair return on their railway property held for and used 
in the service of transportation." Moreover, under the "recapture 
clause," railroads earning over six percent on the value of their prop- 
erty were directed to pay one-half of the excess to the Interstate Com- 
merce Commission, which was then empowered to relend, or purchase 
equipment with, the proceeds to help weak lines — but subject to charg- 
ing an interest rate of six percent. Finally, the Act authorized pooling 
for the first time, when approved by the Commission, and, also for 
the first time, permitted the Interstate Commerce Commission to pre- 
scribe minimum rates. Ancillary provisions which had some bearing 
on the "fair return on fair value" objective included power to the 
Commission to restrict new construction, to require construction of 
new lines, to control abandonments, to prescribe intra-state rates, 
and — on the financial side — to control the issuance of new securities.^^ 

In evaluating tlie operation of the Act of 1920, the verdict with 
respect to its professed main objectives nuist be "non-feasance." But 
the non-feasance cannot be attributed exclusively to the Interstate 
Commerce Commission. It was also due to the characteristics of the 
transport system and, in the 1930s, of the economy as a whole. 

The results of the Act of 1920 do provide glaring examples of regu- 
latory lag. Professor William Z. Ripley, of Harvard, set to work on a 
consolidation plan with zeal, immediately after the passage of the Act, 
and submitted a preliminary consolidation plan to the Commission in 
January, 1921. The plan was largely incorporated into the Commis- 
sion's first tentative consolidation plan.^* Throughout the 1920s, how- 
ever, there ensued just the kind of railroad combination movement 
which one might expect from the 5zeneral financial practices of the 
era — association by holding company, and not by outriglit merger. 
Moreover, the Van Sweringens and other newcomers on the railroad 
scene cut a bold swath through the Eipley Plan and practically every 
official pronouncement. At last — "Pieluctantly, the Commission issued 
its complete plan December 9, 1929." ^^ It had virtually no effect. 

These comments lead us to a summary of the actual consequences of 
the Act of 1920: 

(1) Its net contribution toward unscrambling the Balkanization 
of the American railroad industry was at best zero. One of the reasons 

13 Thi"? description of the Act of 3020 draws heavily on D. Philip Locklln, "Economics 
of Transportation," seventh edition, Richard D. Irwin,' Inc., Homewood, III., 1972, chapter 
11. "The Transportation Act of 1920." pp. 240-54. 

J* "Consolidation of Railroads," 63 I.C.C. 455 (1921). 

" Leonard, op. cit., p. 186. 


for this futility is still important : strong railroads are most reluctant 
to absorb weak railroads purely in the interests of national transpor- 
tation policy. 

(2) Its net contribution toward equalizing competitive strength 
of railroad systems by enforcement of the Recapture Clause may also 
be summarized as — zero. The Recapture Clause had a fatal defect in 
its initial construction which the Commission could do nothing about. 
This may be described as the "fallacy of the guaranteed loan." The 
word "loan" implies some hope of repayment. Hope of repayment 
implies not only marginal earning power on the part of the debtor 
(i.e., that the loan can be put to uses which will yield more than the 
interest cost) , but also average earning power on the part of the debtor 
in cases (e.g.) where one problem is negative cash flow. Presumably 
private lenders are not blind to the profit opportunities which their 
loans may open up. So the guaranteed loan essentially rests on the 
edge of a razor blade : how can it be at once so bad that no private 
lender would make it without a guarantee, and so good that it will 
be repaid ? Therefore, if "weakness" took the form of an average return 
on investment of less than 6 percent, what general hope could there 
be that this weakness could be remedied by loans at 6 percent? 

(3) The objective of "The Rule of Ratemaking" not only did not 
survive the Depression ; it has never been resuscitated since. Locklin, 
after noting that the Cormnission specified 5.75 percent as a fair rate 
of return, then adds this sentence which sheds light on the entire prob- 
lem of the function of regulation in intercity freight transportation r 
"Since then, the Commission has not specified what it would consider 
the appropriate rate of return." He adds, 

... In fact, there has been little occasion for it to do so notwithstandinj^ the 
large number of rate-level cases that have arisen. Increases in rates sought by 
the railroads have generally been less than would be necessary to earn a return 
under any reasonable concept of a fair rate of return. . . ." 

In 1959, Merrill Roberts commented that the Commission actively re- 
jected return as a guide to earnings reasonableness.^^ But this leaves 
the problem which the "Rule of Ratemaking" sought to resolve open 
for consideration in a new form: what kind of rate-of -return policy 
should a regulatory body adopt toward a mature industry, and what 
kind of effects should it expect from this policy in terms of attraction, 
retention, or expulsion of capital? And what if the industry is not 
expected to be able to earn any "suggested" return? 

(4) Perhaps the outstanding example of non-feasance which can be 
attributed to the Interstate Commerce Commission itself, and not to 
the economic environment in which it operated and operates, is that 
pertaining to rate "divisions" (of revenues from joint rates). Here the 
Interstate Commerce Commission was alert at the very outset ; it gave 
special consideration to the revenue needs of New England railways 
by raising their divisions, in 1922, and was upheld by the Supreme 
Court in 1923.i« But much of the northeastern railroad system has since 
deteriorated into a collection of bankrupt enterprises without any gen- 
eral action, from within the Interstate Commerce Commission or with- 
in Quotations are from Loeklin. op. cit. p. 400 and pp. 400-401. But see nre^ient ICC 

proceedings in connection with ex parte 353. 

17 Merrill .T Roberts. ••Maximum Freight Rate Regulation and Railroad Earnings 
Control." Land Economics. 35 (1959), 132. 

TT Q "?oT /?n?t\°^ Divisions," 66 I.C.C. 196 (1922) ; "New England Divisions Case." 261 
U.o. lo4 (19^3). 


out, on the possibility of altering divisions. Such action would, of 
course, raise the question : assuming the rich are expected to give to 
the poor, are there anv rich ? • • r 

The Act of 1920 mav tell us more about the characteristics of reg- 
ulation of railroads bv its failures that it could have told us if it had 
been successful. First' of all, to the extent that railroads are at least 
partiallv competitive, the "local monopoly*' attribute which is the 
foundation-stone of public utility regulation is not so much a rock on 
which to build as a potential quicksand. Eegulation of many partially 
competitive firms in dilf erent industries is much harder than regulation 
of one. Second, a merger or any other policy carried out solely to en- 
able the introduction of full public-utility type regulation would be 
likely to sacrifice a number of truly economic objectives on the altar of 
administrative convenience. Third, an attempt to force or persuade 
independent enterprises to combine for such purely administrative rea- 
sons cannot work on a voluntary basis; it contains no appeal to any 
private profit motive. Fourth, palliatives such as changes in rate di- 
visions can work only if disequilibria are regional and not evidenced 
by parallel or otherwise competing railroads: also, like any other form 
of redistribution, this form requires wherewithal at the source as well 
as need at the destination. 

In sum. the Act was a failure not because of any intellectual de- 
ficiencies on the part of its proponents, but because, by 1920, history 
was already beginning to pass them by.^^ 


So far, the discussion has been entirely about regulation of railroads. 
Part, of the reason for this is jurisdictional: water transport of all 
kinds is excluded from the purview of this paper, unless it specifically 
impinges on rail transport, because it is largely unregulated. But the 
main reason is historical : the main alternative to rail freight is truck- 
ing, and that only became significant in the 1920s. 

The leading Supreme Court cases related to motor carriers which 
resulted in decisions against state regulation were all decided in 1925.^** 
The regulatory vacuum created by the Supreme Court decision was 
first approached through self-regulation, under the X.R.A. codes; 
and, after the Schecter decision, replaced bv the Motor Carrier Act 
of 1935.21 

In view of subsequent developments, the massing of forces with 
respect to this Act is of particular interest. Groups .favoring regula- 
tion included state commissions, the Interstate Commerce Commis- 
sion, the railroads, bus operators, and ''a few of the common-carrier 
truck operators.'' " There was "strong and active opposition" from the 
automobile manufacturers, the organized shippers, and the majority 
of truck operators — particularly the contract carriers.^^ This confixjn- 

i» A lively history of the railroad rate-of -return problem which set the sta?e for the 
Act of 1920 is contained In Albro Martin. "Enterprise Denied : Orlcrlns of the Decline of 
American Railroads, 1897-1917." Columbia University Press. 1971. In ideolo^cal position, 
this is almost the exact reverse of Kolko. But the book does lead one to speculate that 
the Act of 1920 was the best possible railroad legislation, passed about twenty years too 

"Locklln. op. clt., p. 674: James C. Nelson, "The Motor Carrier Act of 1935." Journal 
of Political Economy. 44 (1936), 464. 

*i Nelson, op. clt., pp. 469-470. 

"Ibid., p. 466. 

*» Ibid. 


tation was somewhat confused during the X.R.A. period by the fact 
that trucking organizations were in favor of "self-regulation" under 
the N.I.R.A. codes; the Supreme Court's Schecter decision, which 
wiped these codes away with one stroke, converted the American 
Trucking Associations, Inc., from an opponent to a sponsor of the 

The Act of 1935, which became Part II of the Interstate Commerce 
Act in 1940, is probably more remarkable for the respects in which it 
does not differ from Part I (railroads) than ,for the respects in which 
it does. This Act treated regulated motor carriers as if they were 
railroads in important ways : 

(1) Motor carriers who could establish that they were in business 
for hire, and performing specific services (e.g., between stated points, 
over stated routes, carrying specified commodities), at tlie time of 
the passage of the Act in 1935, were granted "grandfather rights." 
Other carriers who could not establish such prior performance were 
forced to obtain "certificates of convenience and necessity" for com- 
mon carriers ("permits" for contract carriers) for the specific new 
service they wished to render, whether they were already in the 
industry or not. Thus "control of entry" was not introduced in the 
generalized sense that one would expect to result from the technology 
and economics of the trucking industry, but often in the specific sense 
(e.g. for regular route common carriers) appropriate to railroads. 

(2) The provisions o,f the Act with respect to rates, tarifi's, and 
method of price quotation are practically identical for rail and truck. 
This means that, in both cases, the Intei^tate Commerce Commission 
can establish the maximum, minimum, and actual rate; in both cases, 
thirty-days' notice is required for changes in rates, publication of 
rates is required, these rates must be adhered to, and the Commission 
may suspend proposed rates for a period of not to exceed seven 
months. The original tariffs filed by m.otor carrier rate bureaus (ex- 
cept for New England) were taken over bodily from railroad rates. 

No one would contend that present-day motor carrier tariffs are 
tied to those charged by rail. Nevertheless, the original 1935 approach 
Vv'as based on tlie essential proposition that "fair competition" should 
I'equire Mode B to try to assume all possible characteristics of Mode 
A, whetiier or not they were even appropriate to Mode A and utterly 
without reference to the economic characteristics of Mode B. Hence 
a whole series of mutually-aggravating^ anomalies: railroad rates on 
manufactures and miscellaneous tended, in 1935, to be based on value- 
of-service in addition to or in some instances almost in place of cost of 
rail transport; costs of truck transport had nothing to do, in level or 
structure, with costs of rail transport. Thus motor carriers adopted 
a rate structure which was two removes from anj cost justification. 

This anomaly has since been chipped away, not primarily (if at all) 
by regulation, but by competitive forces within the trucking industry 
whicli will be described and anah^zed in the next section. But a further 
anomaly remains, connected not with the level and structure of final 
rates but with the process of rate-making. Any kind of systematic 
railroad rate-making (including systematic favoritism and discrim- 

2* Ibid., pp. 469-470. 


ination) involves a certain amount of inflexibility and hierarchical 
organization. Large and far-flung firms cannot merely charge prices; 
they must have a price policy. None of these conditions was even re- 
motely applicable to the vast majority of trucking firms in 1935 for 
rail-competitive truckload trailic — or to most trucking firms even 
today. Organizationally, they are small enough to be flexible as to 
rates. Economically, they are also sm.all enough so that this flexibility 
could not possibly assist some potential new Standard Oil Trust. But 
the process of ratemaking by motor common carriers is still the process 
designed around the peculiarities of an industry which was first sub- 
ject to regulation 48 years before 1935. Mode B is regulated not only 
as if it were Mode A, but also as if it were Mode A in the complete 
absence of Mode B. 

Against this background of the "as if regulatory similarities of rail 
and motor common carriers, differences in the law are few and grudg- 
ing. These are among the more important : 

(1) Motor carriers must, as railroads need not, carry external lia- 
bility protection. Some of the larger motor carriers are now a good 
deal more solvent than many railroads. But this difference in legal 
standards gives at least some acknowledgement to the vast difference 
in modal size of firm as between rail and motor carrier in 1935. In 
lesser degree, this difference in modal size persists. 

(2) Although the Interstate Commerce Commission has power to 
compel railroads to establish through routes and joint rates, it has no 
such power over motor carriers. This lack of power has provided the 
occasion for the introduction of numerous bills.-^ The problem of 
througli routes and joint rates is one of first importance. Shippers, 
understandably, prefer single-line trucking service to multiple-line 
service."*^ ]\Iotor carriers, understandably, are convinced that profit- 
ability decreases as the number of carriers involved with a particular 
shipment increases.^^ In 196G, joint line revenues ranged from 16 per 
cent of the total for the rate bureau reporting the lowest proportion 
to 51 per cent for the highest two.-® But with the consent, if not at 
the insistence, of the Interstate Commerce Commission, rate divisions 
have been based on railroad division scales."^ 

In terms of operating revenues, motor common carriage of general 
com.moditios is largely a less-than-truckload business. Less-than-truck- 
load transportation is likely to involve physical transfers. If a ship- 
ment passes from carrier to carrier, the number of these transfers is 
likely to increase. So the present system, by its reliance on operating 
rights, gives a differential advantage to concerns who already have the 
most extensive systems, and may impose penalties on a carrier with 
less extensive operating rights who would still like to try to overcome 
his differential disadvantages and compete for long-haul business. 

*5 A*", one example, see "Thron<rh Routes and Joint Rate?." hearlnfrf? before the Pub- 
cnnnnlttee on Surface Transportation of the Committee on Commerce, U.S. Senate, 91st 
Conjrress. 2rl Session, on S. 2245 and S. 3G26. G.P.O.. Wa'^binjrton. D.C. 1970. 

"See Jnmes C. Johnson. "An Analysis of the ICC's Adr^lnlstratlon of Section 5 Trucking- 
Mergers." Prof'eedlnprs — Fourteenth Annual Meeting. Transportation Re.=:enrch Fornm, 
V. 14. Xo. 1. 107.'^. Richard B. Cross Co., Oxford, Ind., 1973, pp. 773-793. especially 
pafres 774-77.5 and 781. 

*^ Gerald E. Hawkes. "Motor Carrier Divisions of Revenue on Joint Rates Traffic,'^ 
Tr:in<;''ortntlon JnurnMl. vol. 8, Xo. 1 (fall 1968), table 4, p. 28. 

«Thid.. tnhle .S. p. 27. 

" Ibid., especially p. 31. 


Current Regulation 
what freight transport regulation is not 

It is not effective control of each funn's rate of return 

As should already be clear, all railroad regulation lacks one dimen- 
sion of public utility regulation : it is not concerned, in practice or even 
in principle, with return on investment. ^^^ There are two reasons for 

The first has already been discussed in connection with the Act of 
1920. But it may be summarized briefly here. Railroading is an in- 
dustry all of whose competing enterprises must charge the same prices 
from the same origins to the same destinations, and in which more-or- 
less competitive firms have different historical backgrounds, different 
cost levels and structures, different marketing skills, and different 
complementary and competitive relationships with other elements in 
the system beyond the relationships which exist on the route from 
A to B. Therefore it is impossible to establish any target rate of re- 
turn which all enterprises have an equal chance of attaining. More- 
over, in an industry which is geographical in its very essence, differ- 
ential rates of change in economic conditions in different regions will 
tend to shift relative rates of return over time even if they were equal 
at the outset. 

The second reason is that neither the railroad industry nor the 
Interstate Commerce Commission even thinks in terms of the kind 
of argument, or the kind of rate of return, which are staples in rate 
cases involving telephone or electric power companies. Underlying 
this fundamental reason is an obvious doubt, on the part of all con- 
cerned, as to the prospects of the railroad industry as a whole of earn- 
ing what a utility would consider a "normal" rate of return. 

Both of these difficulties have important economic consequences, 
which will be discussed later. It should be remembered that the very 
idea of "fair return on fair value" or "on original cost," or on any 
other basis, has disappeared from railroad economics, except when its 
opposite — "inadequate return" — is cited as a reason for granting a rate 

There are, of course, persuasive arguments against attempting to 
give the railroad industry a normal rate of return on its investment. 
The obvious one has already been mentioned : why try, when it is prob- 
ably impossible? This in turn is associated with the allegation that 
much railroad investment is obsolete, if it was even justified in the 
first place, and kept on the books only because the accounting prin- 
ciples used for railroad way and structure are different from the ac- 
counting principles used for all other fixed assets in the economy. Fi- 
nally, there is a combination of "sunk cost" and "senescent industry" 
arguments : railroad fixed capital is there ; much of it need not or, eco- 
nomically speaking, should not, be replaced; why not get the maxi- 
mum use out of it in the meantime by pricing its output on a basis 
which excludes capital costs ? 

These general arguments do not hold for regulated motor carriers. 
By anyone's standards, this has been and is a growth industry: no 
one suggests that the capital investment should be gradually squeezed 
out of it. The "way and structure" portion of its growth has been 

29a The 4-R Act of 1976 has created the occasion for ex parte 353 — ^which addresses 
this problem in principle, if not in practice. 


largely created bv very extensive governmentally-financed road pro- 
grams. The railroad industry has always approached public utility 
status in its ratio of capital investment to current reveunes, which 
even with recent precipitous rate increases is still running close to 2 :1. 
The trucking industry, on the other hand, has a ratio of capital in- 
vestment to operating revenues which places it somewhere between 
manufacturing and grocery chains. It also has many more competitive 
firms on most major routes than the railroad industry. So the regu- 
latory problem with the common carrier trucking industry is not 
whether it should earn a normal return. In view of the growth of cap- 
ital assets in the industry pari passu with its growth in revenues, the 
pragmatic answer has obviously been, "Yes, a normal return at least:'' 
Xon obviouslv, is the question whether it can earn a normal return. 
The real question is, '*AVhat should, and can, the Interstate Commerce 
Commission do about all this?" And an attempt to answer this ques- 
tion leads one from the mechanics of ''fair return on fair value" into 
the underlving question — whicli is not necessarily "Should common 
carriers by motor be regulated?", but "With respect to rate of return, 
is there any economically-meaningful and administratively-practi- 
ca])le basis on which they can be regulated ?" 

To summarize this section, freight transport regulation is not rate- 
of -return regulation, in any way, for railroads. Its status in the truck- 
ing industry remains ambiguous. The problem with respect to motor 
carriers is not Interstate Commerce Commission recognition of rate 
of return as a criterion. Its administrative aspect is: how, with a 
capital-revenue ratio of 1:3 or 1 : 4, and with as many as fifty or so 
competitors on the busiest routes, can any administrative body make 
decisions which are relevant to particular carriers? From a purely 
administrative standpoint, can any meaningful criteria be developed 
for appropriate rates of return on capital in an industry with a high 
rate of capital turnover, a high rate of national unionization, and a 
high ratio of direct wages to operating revenues? 

Some of the dilferences which have just been described are presented 
in quantitative form in Table 1 below, for pre-depression 1973. 


[Dollar amounts in millionsl 

Line-haul Motor 

railroads carriers 

Operating revenues... $14,990 $16,478 

Net transportation operating income 698 866 

Ordinary income ' 401 483 

Net income 451 495 

Federal income taxes on ordinary income 113 266 

Net investment in transportation property 27,699 3,565 

Shareholders' and proprietors' equity 16,422 3, 170 

Operating ratio 79.38 94.75 

Return on net investment (percent) ■ 2.52 16.82 

Return on equity: Ordinary income basis 2.44 15.62 

> "Ordinary income" is derived from "Net transportation operating income" by adding income from nontransportation 
sources, and by subtracting interest or other payments attributable to the acquisition or use of capital assets, and income 

Source: Interstate Commerce Commission, 88th Annual Report to Congress, 1974, Washington, D.C., GPO, 1974, appen- 
dix E, tables 6 and 10, pp. 122 and 125. 

(Data are for 1978 Instead of for 1975 — the most recent year for which data are avail- 
able — because net railway operating income was negligible, and railway ordinary income 
negative, in the latter year.) 


Thus the railroads had a capital-turnover ratio, as measured by the 
ratio of operating revenues to net investment in transportation prop- 
erty, of 0.54 ; for motor carriers, the comparable ratio was 4.62, or 8.6 
times as great. Plence the relatively minor advantage of motor car- 
riers over' railroads in operating income from transportation activities 
($866 million vs. $698 million) , was translated into an enormous differ- 
ential in return on investment in transportation — 2.52% for railroads; 
16.82% for motor carriers. This, in turn, was by far the most important 
single influence in determining final percentage returns on equity of 
2.44% and 15.62%, respectively. The 15.62% return reported by Class 
I motor carriers compares with 12.8% reported for all manufacturing 
corporations in 1973, and was higher than any return for manufac- 
turing corporations after 1948.^° The I.C.C. 90th Annual Eeport to 
Congress (1976) shows return on equity for Class I motor carriers at 
more than 13% for each of the vears 1973, 1974, and 1975: railroad 
returns for these years were 2.44%, 3.07%), and a deficit of $196,157,000 
(Appendix E, Tables 6 and 10) . 

It is mot complete control over finance; and it is in only an indirect 
sense control over investment 

The Interstate Commerce Commission has had control over certain 
forms of railroad financing since 1920. and of certain forms of truck 
financing since 1935. But. in both industries, leasing and other ways of 
shifting portions of the finance of the carriers tovrard outside parties 
are common. In the motor carrier industry a substantial fraction of 
total assets is provided under leasing and other arrangements witji 
owner operators. These arrangements are in some cases both non-arm's 
length and beyond the powders of financial control of the Interstate 
Commerce Commission. So, even in the narrowly financial sense, the 
Interstate Commerce Commission's control is considerably less than 

Proceeding from new fi.nance to the kinds of investment that mio:ht 
make such new finance necessary : for both rail and motor carriers, 
inventories are negligible and accounts receivable are quite strictly 
limited bv Interstate Comm.erce Commission rejrulations aim.ed nt 
diminishing the indii-ect rate discrimination which mis^ht otherwise 
occur if some shippers were granted m.ore lenient credit terms than 

As for fixed assets, both the industries and the regulatory ruies 
under which they operate differ because of the fact that railroads 
must provide their own way and structure whereas motor carriers in 
effect "lease" the rights-of-way and roadbeds they employ through tlie 
payment of various vehicle and fuel taxes. Thus the Interstate Com- 
merce Commxission has had. since 1920. authority over new railroad 
construction as well as abandonments which it has not had with respect 
to motor carriers. But the new construction authority has been of 
minimal significance, for the obvious reason that the railroad network 
has been quite steadily contracting since 1920. The onlv maior now 
construction which is now contemplated is in connection with t:.e 
opening-up of coal mininc: areas in the northern Great Plains. Prnc- 
tically, then, I.C.C. control over investment in transportation (whether 
net or gross) does not even extend to control over replacement. 

30 Econnrnic Report of the President, January 1976, G.P.O., Washington, D.C., 1976, 
App. B, Table B-76, p. 261. 


This absence of control over practically all the important aspects 
of the investment policy of carriers (with the notable exception of the 
disinvestment policy represented by railroad abandonment), has 
necessarily had repercussions on various other features of supposed 
Interstate Commerce Commission control : 

(a) In the trucking industry, which has been expanding rapidly 
and almost continuously since 1945. I.C.C. inability to control entry 
of investment raises serious questions as to the significance of I.C.C. 
regulation of entry of carriers, as well as to the significance of I.C.C. 
regulation of service and service standards. 

Any regulatory influence on minimum rate levels must, if it is to be 
practically meaningful (as all parties concerned claim that it is), 
establish at least some rates which are above the rates which would 
be tlie result of free bargaining between carriers and shippers in an 
unregulated market. Given tlie existence of some such rates, and given 
also the inabilit}^ of the Interstate Commerce Commission to enforce 
miy particular standard of service in situations where legally-imposed 
m.aximum rates might otherwise result in losses to the carrier, there 
must be a tendency to create excess capacity, or excess service, in the 
most competitive portions of the regulated motor carrier industry with 
respect to particular shippers, shipments, commodities, or routes. This 
may (or may not) take the form of excessive empty back-hauls. It may 
also take the form of partial loads where full loads might otherwise 
have been obtained, excessive delays of carrier personnel and equip- 
ment at loading and unloading docks, extra freight solicitation, etc.^^* 
But. whatever form it takes, the consequence of an imposition of effec- 
tive minimum rates in competitive situations is bound to be the crea- 
tion of excess physical capacity, or excess service effort, or both. Rate 
formation through rate Inireaus re-enforces these inherent tendencies. 

Tlicse inevitable i-esults of any minimum rate regulation which is 
economically effective (i.e., which resuhs in prices to at least some 
ship))ers liigher than they would otherwise be), is exacerbated by the 
peculiarities of American regulatory law and practice. American 
VAotov carrier regiilation for regular route common carriers of general 
comniodities operates route-by-route, and sometimes gateway-by-gate- 
way. The Interstate Commerce Conmiission has shown itself willing on 
more than one occasion (the appearance of the Interstate Plighwav 
System: the so-called -'energy crisis" which began in October 1973) 
to permit regulated motor carriers to take advantage of new routes 
vrhich will enable them to serve their customers more cheaply as well 
as more expeditiously. But, in both of these important new economic 
situations, the Interstate Commerce Commission has drawn the line 
at permitting any carrier to take advantage of the relaxation of pre- 
vious rules if his saving in route mileage would be more than a pre- 
scribed standard. Thus, if placed in a genuinely new economic environ- 
ment, the Interstate Commerce Commission tends to straddle on the 
subject of gams m efficiency: up to a point, such gains are to be en- 

of private carriers 


couraged ; but it is assumed that major gains by any one carrier might 
upset the competitive balance on individual routes, and are therefore 
to be prevented. 

So, (1) effective minimum prices in a genuinely competitive environ- 
ment necessarily create excess service or excess capacity; but (2) in 
the context of American motor carrier regulation, efforts to hinder the 
development of a genuinely competitive environment tend to freeze 
the worst-situated potential competitors in their actual situations of 
substantial inferiority on particular routes. (3) Meanwhile, better- 
situated firms now in the industry are in a position either to receive 
excess profits as a result of their favorable operating rights, or to sell 
some of these "intangibles" at prices which capitalize their quasi- 
monopoly value. 

(h) In the railroad industry. Interstate Commerce Commission con- 
trol of track abandonment, when combined with the absence of Inter- 
state Commerce Commission control of the condition of track in use 
(powers to establish track standards are vested in the Federal Rail- 
road Administration), creates a tendency toward exactly what can. be 
observed almost throughout the ISTortheast and in much of the Mid- 
west — a maximum of ramshackle mileage. The kind of service and 
accident record which much of the railroad industry has been com- 
piling is an obvious product of regulatory lack of interest in what, 
in the circumstances, might constitute an "adequate" rate of return, 
plus quite severe regulatory standards with respect to one form of 
disinvestment (line abandonment), plus minimal regulatory controls 
over other forms of railroad investment or disinvestment. 

(c) Also in the railroad industry, new stock issues are almost un- 
known and "new" bond issues tend to be limited to roll-overs of old 
issues as they mature. Equipment has been financed by a number of 
measures, most of which require only partial or zero cash down-pay- 
ment. The flow of funds to railroad way and structure is constrained 
by two facts : way and structure is immobile, geographicallv, and often 
further immobilized legally by "after-acquired" clauses m bond in- 
dentures which commit any new investment in way and structure as 
further security for the bond holders who have first claim against 
railroad assets. But internal investments in way and structure compete 
with dividends ; and both compete with use of railroad cash flows for 
diversification purposes. 

li does not, according to the Interstate Commerce Commission, exercise 
more than very limited de facto control over changes in individual 
rail rates 

There can be no question about the Interstate Commerce Commis- 
sion's legal powers over rates charged by railroads or regulated motor 
common carriers. The I.C.C. can set them on its own motion; it can 
suspend proposed changes, investigate them, and then approve or 
deny them; it can establish maximum, minimum, or actual rates. 

The important question is not what the Commission may do if it 
so desires, but what it actually does. And, on this point, the Commis- 
sion s claims are exceedingly modest. For example, here are the an- 
swers furni-hed by George M. Stafford, former Chairman of the 
Interstate Commerce Commission, to Representative Fred B. Rooney, 


Chairman of the Subcommittee on Transportation and Commerce of 
the House Committee on Interstate and Foreign Commerce : 

Question 1. In your testimony of July 16, 1975, you indicated that the Inter- 
state Commerce Commission has been very liberal in allowing railroad rate 
reductions to go into effect. Would you please supply a breakdown of the total 
number of such reductions applied for since 1970 and the total number 
suspended ? 

Answer. For fiscal years 1971 through 1974 there was 1,163 rail tariff publi- 
cations containing rate reductions considered for suspension. Of these 149 
publications w^ere suspended in 61 Investigation and Suspension proceedings 
and 1,014 were not suspended. During this four-year period there were an 
additional 255,838 rail tariff publications containing all types of changes which 
were permitted to become effective. 

* ♦ * • • ♦ ♦ 

Question 7. With respect to rate increases, do you have any figures on how 
many rate changes since 1970 were applied for, and the total number that were 

Answer. For fiscal years 1971 through 1974 there were 2,081 rail tariff publi- 
cations (not including ex parte general increases) containing rate increases 
considered for suspension. Of these 1,271 were suspended and 810 were not 
suspended. During this four-year period there were an additional 254,920 rail 
tariff publications containing all types of changes which were permitted to 
become effective.^^ 

These responses raise so many new questions that consideration of 
most of them must be deforred to Jater sections. But a .few additional 
questions min^ht be mentioned in passing: 

First, there is the question of relative importance. Former Chairman 
Stafford listed nine rate suspensions which involved proposed rates 
which were below the Interstate Commerce Commission determina- 
tion of variable cost, out of 62 suspensions.^^ These nine involved the 
following : 

Eates on paper in trailer-on-flatcar service. 
Eates on glassware in trailor-on-flatcar service. 
Kates on popcorn, for export (Sub. 1.). 

INIultiple-car rates on freijjht, all kinds, in trailer-on-flatcar 
service (T. & S. Docket No. R7T7) . 

]\rultiple-car rates on freight, all kinds, in trailer-on-flatcar 
service (I. & S. Docket ^o. 8787). 

Intercoastal-water-competitive rates on chemicals. 
Intercoastal-water-competitive rates on cliemicals [Docket No. 
8840. Since the previous entry contains no docket number, it may 
represent an accidental duplication]. 

Multiple-car aggregate rates on wheat, in bulk. 
Proportional rates on grain and grain products.^^ 
Barring a revolution in world eating habits, the export popcorn 
rate suspension will probably not determine the fate of the railroads, 
the popcorn-growing industry, or the Interstate Commerce Commis- 
sion. But the other cases clearly involve commodities, and movements, 
of considerable potential importance. 

31 "Railroad Revitalization," hearins:s before the Subcommittee on Transportation and 
Comniorro of the romniittee on Intervtate :'nf1 Fore1<rn rnniT^ipr'^o. Hcn«e of Renrpsentn- 
tires. n4th Concress, 1st Session on H.R. 6351 and H.R. 76S1, Serial No. 94-38, G.P.O., 
Wnqhlnrrton. D.C.. 107o. pp. .•^52. 355. 

32 On. cit.. \\x). 353-354. This number of 02 presumably differs from the 61 previously 
mentioned because one of the suspensions in his list has no docket number. 

"Loc. clt. 


Second, for the upshot of all this: 

Question 4. How many of the rate reductions that the Commission suspended 
were ultimately found unlawful? 

Answer. A number of the suspended rate reductions . . . were not disposed 
of by final report. Twenty-eight of those proceedings, for example, were dis- 
continued after the carriers voluntarily cancelled the schedules at issue with- 
out defending them. In eight other instances, the carriers failed to submit any 
evidence and the schedules were ordered canceled. The Commission acted favor- 
ably upon petitions to vacate the suspension in seven cases and the proceedings 
were discontinued. The schedules were ultimately found unlawful or not shown 
to be just and reasonable in eight cases. Five of the proceedings . . . are still 

These questions and answers all relate to railroad tariffs, but the 
o:eneral picture of a very low ratio of suspension to new rate filings 
would be maintained if motor carriers were included. For example, in 
fiscal 1974 73,233 rail tariffs were received, along with 236,907 motor 
common carrier and 20,768 motor contract carrier tariffs, for a com- 
bined total of 330,908. These tariff submissions constituted 86.8 percent 
of all freight tariffs filed with the Interstate Commerce Commission 
during that period. ^^ During this same fiscal year, 452 rail and 3,205 
motor carrier rates were considered for suspension (equal to 0.62 per- 
cent and 1.24 percent, respectively, of filings during the year), and 
232 and 1,521, respectively, were suspended in whole or in part (equal 
to 0.32 percent and 0.59 percent) .^^ So, although the rate of motor car- 
rier suspensions was practically twice as high, these were still a trivial 
percentage of rate filings. 

If all of these numbers are taken to prove that the Interstate Com- 
merce Commission really interferes hardly at all with competitive 
market processes, then the consequences for any case for regulation 
are quite evidently disastrous: if the Interstate Commerce Commis- 
sion does not even exercise any control over rates, how can it justify 
its annual budget? The numbers are, in fact, a quantitative example 
of the kind of blowing hot and cold wliich is typical of discussions of 
regulatory matters. The effects of regulation can be greatly exag- 
gerated for some purposes, or to some audiences, while simultaneously 
being deprecated for other purposes, or before other groups. 

But the modesty of the Interstate Commerce Commission with re- 
spect to its use of investigation and suspension powers becomes it in 
at least two respects. It lays to rest a kind of paranoid vision of au- 
thoritarianism which is based on the false proposition that "the Inter- 
state Commerce Commission sets transportation rates." Correlatively, 
it brings out the inherent passivity of the Interstate Commerce Com- 
mission (and, generically, of all "economic regulation" contructed on 
the I.C.C. model). 

It is losing some degree of control over the qyxtntity of intercity 
freight transportation^ whether measured in the aggregate^ inter- 
modally^or within each mode 

(a) In the aggregate 
Table 2 provides a summary of the "horizontal," or modal, extent 
of regulation. 

s*Ibid., p. 355. 

"I.C.C, 88th Annual Report to Congress. 1974, Table 5, p. 108. 

»« Op. cit., tables 5 and 6, pp. 108 and 109. 



[In billions of ton-miles] 

Federally regulated 
Ton-miles Percent 
































2, 439. 7 


1 Only regulated traffic carried by regulated water carriers. 

Source: I.C.C. 90th Annual Report to Congress, 1976, app. E, table 4, p. 143. 

If this table were modified by insertin*! value and not volume of 
transportation, as well as degree of regulation, the 63.5 percent shown 
in the table Avould doubtless appear to be too high. Oil pipelines come 
close to being plant facilities, and are not regulated in the same way 
and to the same degree as rail and motor common carriers. The motor 
carrier segment of the intercity freight transportation industry ob- 
tains much higher receipts per ton-mile than the railroad industry 
(although water carriers have lovrer receipts per ton mile). 

In any event, the 63.5 percent shown in Table 2 is much lower than 
it would have been in 1920, or 1945. or mo.^t other years up to the very 
recent past. Eailroads are completely under Federal regulation. Their 
more rapidly-<rro\ving competitors either contain a major non-reg- 
iiUited component (iiiotor and water), or are subjected to quite diller- 
ent forms of regulation as a partial reflection of their quite different 
technical, instituticnal, and economic characteristics (e.g., oil 

Therefore, Federal regulation is steadily losing scope: 

(h) Intei^modalhj 
As the preceding section should have made clear, the global decline 
in the quantitative coverage of regulation has been a result of modal 
shifts toward less-regulated forms of transportation. 

{c) Intramodally 

This question cannot be answered by a simple appeal to ton-miles. 
]\roreover, it not only has the horizontal dimension (''Federallv reo;- 
ulated"' V. "Xot federally regulated") as presented in Table 2; it also 
has a vertical dimension. 

Horizontal measures of the importance of regulation. — In this re- 
spect, Table 2 is doubly at fault— for reasons that are opposed to each 
other, but do not cancel out. 

Table 2 exaggerates the economic significance of Federal regulation 
bv using ton-miles as the criterion rather than operatincr revenues. 
Oil pipelines liave far lower revenues per ton-mile than motor carriei-s, 
and therefore far lower total revenues; vet. on a ton-mile basis. Table 
2 gives them almost twice the weight of motor carriers. Class I rail- 
roads, which are 100 percent Federallv-regulated, had average freight 
revenues m 1974 of 1.8480 per ton-mile; Class I intercity motor com- 
mon carriers had average revenue of 9.000^, and Class I contract ear- 
ners had average revenue of T.OOO^.^" 

^" "Tran.sp.)rt Economics,"' vol. Ill, No. 2 (1976), p. 14. 
S3-944— 78 3 


This exaggeration is also related to tlie danger of exaggerating the 
degree of direct competition between railroads and regulated motor 
carriers. Naive intruders into the world of transportation have been 
known to wonder how an industry with an average revenue of 9 cents 
per ton-mile can possibly be a stern competitor of an industry with an 
average revenue of 1.8480. The answer is just what these statistics 
would lead one to suspect: the two regulated industries do largely dif- 
ferent things, in different ways, to satisfy different demands. For exam- 
ple, by far the largest Class I motor common carrier (considered as an 
integrated holding company, not as a group of operating subsidiaries) 
is United Parcel Services. Yet United Parcel Services is an extremely 
specialized carrier : it will accept no single package which weighs more 
than fifty pounds, and cannot accept combined shipments from one 
source, to one destination, at one time, of more than three packages 
weighing more than a total of one hundred pounds. Obviously any- 
thing that is done in connection with regulating U.P.S. is quite irrele- 
vant to railroad regulation, and only marginally relevant to many 
features of general motor carrier regulation. 

The same comment holds, in lesser degree, for much of the regulated 
motor carrier industry. In 1972, for example. Class I I.C.C.-regulated 
motor common carriers received $5.4 billion in operating revenues from 
less-than-truckload freight. ^"VTien U.P.S. intercity operating revenues 
are added to this, the new total becomes approximately $6.1 billion.^^ 
For the same year, "Operating revenues — Freight-intercity-common 
carrier" came to $13,564 million.^^ Although the two classifications are 
not strictly comparable, it would not be unreasonable to place less- 
than-truckload revenues for all intercity motor common carriers of 
freight (including specialized, bulk, and irregular-route carriers) at 
close to 45 percent of the total. This high percentage is in face of the 
fact that "less-than-truckload," for this purpose, is determined on a 
tariff and not a physical basis : a single trailer with a twenty-ton capac- 
ity is capable of transporting up to^four "tariff" truckloads of a mini- 
mum of 10,000 pounds each. (For 978 common carriers of general 
freight, in 1976, operating revenues from carriage in less-than-truck- 
load lots (L.T.L.) came to $7,328 million, as compared with $4,125 
million in operating revenues obtained from truckload lots).^^* 

So, measured horizontally, it is clear that much of the revenue ob- 
tained from I.C.C.-regulated motor common carriers is derived from 
business which railroads have largely abandoned. In 1972, for exam- 
ple, all Class I and II railroad revenues from less-than-carload freight 
came to just over $28 million — approximately 0.15 percent of all rail- 
road operating revenues, and 0.5 percent of Class I and II motor less- 
than-truckload shipments. 

Thus, on a ton-mile basis, the importance of I.C.C. regulation is 
exaggerated to the extent that, as compared with motor carriers, rail- 
roads are both 100 percent regulated and have much lower revenues 
per ton-mile. To the extent that I.C.C.-regulated motor carriers tend 
to specialize in less-than-truckload shipments, as compared with pri- 
vate carriers and carriers of exempt commodities, then the ton-mile 
measure both exaggerates the importance of private and exempt truck- 

38 "Transport Economics?," vol, I, No, 3 (1974), p. 12. 
3» I.C.C. . ,SSth Annual Renort. App. E, Table 10. p. 12o. 
39a Trinc's Blue Book of the Trucking Industry, 1977 edition, p. 5-3, 


ing within the motor carrier industry and also esago-erates the degree 
of competition between railroads and regulated trucking. 

Vertical measures of the importance of regulatioit. — By "vertical" 
is meant the degree to which the regulated carrier performs the entire 
transportation function for which he is paid with his own resources. 
In tlie railroad industry, the effective answer is "100 percent,-' with 
respect to operations; even new capital investments such as freight 
cars may be the product of railroad shops and not of outside purchase. 
In the motor carrier industry, there is only one important example of 
this kind of vertical integration: Consolidated Freightways builds 
equipment as well as operating it. But many motor carriers do not 
internally provide all the factors of production they need. They may 
lease equipment, witli or without drivers. 

For purposes of this discussion, lease with driver is of greatest inter- 
est, because lease with driver opens the regulated motor-carrier indus- 
try to a completely unregulated entrepreneur: the ownor-oporator.^* 

The most extensive recent study of owner-operators estimates their 
nmnbers at about 100,000, and continues : 

Even this figure, wliich is among the lowest of the estimates we have observed, 
implies an industry generating between $4 and .?G billion per year. . . ," 

From the railroad standpoint, these figures are indeed significant; for 
they would compare witli railwav freiglit revenues of $15.4 billion in 

Although some 40.000 owner-operators are primarily engaged in 
hauling exempt (agricultural) commodities, 58.000 were leased to 
Class I and Class IT regulated caniers in 1072. The power units rep- 
resented in this total comprised 9.10 percent of total power units of 
Class I carriers of general commodities in that year. Owner-operators 
were significantly more important for various categories of special- 
commodity carriage, includimr refrigerated products, building mate- 
rials, and petroleum products.^^ 

The economic significance of tliis availability of owner-operators to 
perform the line haul, is as follows: first, the owner-operator is a 
truck-load carrier: ^^ second, as an independent businessman he is 
essentially not amenable to standard forms of union organization in- 
cluding standard provisions with respect to wages, hours, etc.;** 
third, as a member of a large and obviously mobile industry composed 
of entrepreneurs of his own size and type,' he is likely to bid for busi- 
ness according to the classic textbook model of pure competition; 
fourth, even though he may be only a sub-contractor for carriers who 
hold operating riglits and therefore may seem to have the wliip hand 
(and agree on minimum rates as members of rate bureaus), many of 
the regulated carriers function on highly-competitive routes which 
cannot help but be affected, over the long period, by the price pressure 
exerted by owner-operators; fifth, this pressure is likely to be most 
extreme on high-density motor carrier routes— which tend also to be 
high-density rail routes. 

n.Z?.^^'' i^n^'n ^"^ the extent that the owner-operator is a subcontractor for common 

«n tSoI ,-w'"r"'JJ'''^'?,°^''"^7/'^?^'"^^ ^^th shippers still applv. 

Ty^T^::^'^:^^^^^^^^^^^!^;'^ owner-operator: IndepenSe^nt Trucker/> 

" VySo^iys/e;^^rcii:Sp"i^%'?f • ^^^^"'^^ ^' ^^^^^^ '• ^- ^''- 

" Ibid., p. 8. , , . 

« Ibid., Chapter 5, "Attempts to Organize," pp. 69-83. 


From the standpoint of the legal relationship of owner-operators 
to Part II of the Interstate Commerce Act, the Interstate Commerce 
Commission has been aware of the situation for a long time. For ex- 
ample, a report issued in 1960 comments that 

It is noteworthy that the owner-operator is one of the most important entities 
in gray area trucking operations. . . . 

The owner-operator frequently controls but a single vehicle. . . . Owner-opera- 
tors play an important role in the national transportation system, but many are 
involved in unauthorized operations to some extent." 

In fact, an I.C.C. Summary of "gray area trucking operations" in 
1959 shows "Shipper lease of vehicle with driver" as responsible for an 
average of 44.82 percent of all unauthorized traffic in the regions 

What regulation is not: Conclusion 

The Interstate Commerce Commission does not, and quite probably 
cannot, regulate railroad rates of return on anything like a public 
utility basis ; as for trucking, the potential earning power is obviously 
present, but the "monopoly" and "high capital-output" features which 
are important in utility regulation are non-existent for motor carriers. 
Hence, in the railroad case, rate-of -return regulation is probably im- 
possible economically ; in the trucking case, it at least introduces serious 
additional administrative problems.*^ 

When criteria for earning power are necessarily absent, or very 
difficult to apply to individual firms, control over finance is not likely 
to be very meaningful. In the trucking industry, the Interstate Com- 
merce Commission does not even control leasing and similar arrange- 
ments wdiich are arrived at as the result of non-arm's length bargain- 
ing. In the railroad industry, the Interstate Commerce Commission 
could not exercise effective control over finance without something it 
is most unlikely to be granted---a veto over cash dividend payments. 

What holds for finance holds, a fortiori, for investments. A common 
method of regulation of the motor carrier industry in Western Euro- 
pean countries is to control the capacity owned by common carriers, 
and hence presumably their investment. But, in the United States, 
entry control has nothing whatever to do with investment. It is, 
instead, tied to the particularities of geography or of commodity 
characteristics. In railroading, the only effective form of I.C.C. in- 
vestment control has been negative in character : control of abandon- 
ments. This is typical of the regulatory dilemma. It is relatively easy 
to tell someone what not to do, but it is next to impossible to be an 
affirmative influence on investment policy. 

With respect to control of individual railroad (and motor carrier) 
rates, the Interstate Commerce Commission may perhaps be accused 
of undue modesty. In any event, the consistent position of Commis- 
sioners over the years has been that only a negligible number of rate 
changes are suspended, that many of these suspensions finally result 

*« I.C.C, Bureau of Transport Econonilrs and Statistirs. "Orav Aren of Transportation 
Operations. " G.P.O., Washington, D.C., June 1960 (Statement No, 6010), p. 9. 

*-Ihid., p. 11. 

*^.An article by Harvey A. Levine and Nai Chi Wanff. "Motor Carrier Financing and 
Earnings Regulation: The Other Side of the Coin" (I.C.C, Practitioners' Journal, vol. 42 
(November-Dereinber 1974), pp. 26-41) makes a plea for a more flexible aporoach to 
eirnings regulation of motor carriers than that customarily employed in the utility 
industry. In my judgment, the plea could easily be extended to a plea 'for a more flexible 
approach to motor carrier regulation as a whole. 


in permitting the questioned rate to become effective, and that Com- 
mission objection to a railroad rate Avhich is above the Commission's 
determination of variable cost is almost unheard of. Commission 
spokesmen are certainly right in trying to disabuse the public of the 
notion that the Commission typically sets rates for regulated carriers 
on its own motion. Tliey are probably also right in their implication 
that the Commission is simply not active enough with respect to rates 
to be a reasonably efficient cartel manager. But their further claims 
of inactivity must be regarded with considerable skepticism. 

Once attention is shifted from the substance of regulation to its 
scope in terms of the relative importance of regulated firms, it becomes 
obvious that the only industry which is entirely subject to regulation 
is barely holding its own in freight movements. For motor carriage, 
a rapidly-growing form of transportation, the regulated percentage 
slips to a nuich lower value. The degree of regulation of even regulated 
motor carriers is less de facto, clue to the indirect effects of unregulated 
competition even on regulated carriers. 


TIlis section will consider the positive effects of regulation under 
the following headings: (1) regulation as a means of delaying earn- 
ings increases of regulated carriers; (2) regulation as a means of es- 
tablishing or retaining relationships between different rates charged 
either by different modes or by competitive enterprises representing 
the same mode; (3) regulation as a means of controlling the orga- 
nizational structure of the industry, by entry control and by control of 
mergers; (4) regulation as a means of controlling service standards.. 

Rerfulatwn as a means of delaying earnhigs increases 

The effects of such delay — if, indeed, it exists — are not apparent 
from the profit statistics for i-egulated trucking companies. Their bad 
years tend to coincide with bad business yeai-s in genei-al, or with vears 
of Teamsters' Union sti'ikes. Their worst years tend to occur when 
these two separate influences coincide, as they did in 1970.'*^ 
Xot so for the railroads. 

For this industry, i\\o. Interstate Commerce Commission has made 
very impressive claims concerning the effect of regulation in holding 
down rates. For example, in 1070. Commissioner 'Wal rath presented 
the followino; statistics to the Subcommittee on Surface Transporta- 
tion of the Senate Commerce Committee : 

Between 1051 nnd IDCO : Percent 

Railroad wa,cre rates up 111 

Consumer prices ^p 4]^ ^ 

Railroad materials and supplies I up 86' 8 

Wholesale prices ^^p jf 

Railroad fuel ^ ' -^ -^ 

Carload freight rates'" up 9.' 8 

4.tnTIVHn,re^'?*''''''V I'ruf'l^in? Tronds." 1074. a rnhlirntion of \ho Amrrknn TrMokin- 
inTp^pU ,!;»";• ^.^r""-" ^^'- F* that profits of just over a thousand Class I and Class II 

fell hplOW nprnonf [''^^ ^"^l"'^^ ^'''^'^'"^ '' ''^''■''^' ^O XVOrth" TMtiO wluch 

fnnofhpr^'inhn^ J»^?^ In Only three years between 1963 and 1972 : 1970, 6 percent, 1967 

(another labor contract" year). 8 percent, and 1969. 9 percent. 

GPO W vhin^rn'^DC /.rJ'''''-" ^^-^i-^^-ht. hearings . . ., 91st Congress, 2nd Session. 


Commissioner Walrath added that, over the period 1961-69, rail- 
road wage rates were up 38 percent, raih^oad materials and supplies 
were up 12 percent, and railroad fuel up to 6.2 percent; yet carload 
freight rates were down by 2 percent.^^ 

Since 1969, the comparison would not be so favorable. With 1969 as 
a base of 100, the railroad freight price index was 166.20 in May 1975, 
as compared with the wholesale price index of 162.50.^^ 

Meanwhile, although the Interstate Commerce Commission had 
shifted the basis of its "consumer saving argument," it still strongly 
espoused the argument iself. To quote former Chairman Stafford of 
the Interstate Commerce Commission, testifying in July 1975 : 

Where the benefits or (.97c) regulation can be quantified, the results are im- 
pressive. For instance, had the rates which the railroads requested in three recent 
general rate increase proceedings, ex parte Nos. 265-267, 281 and 295, become 
effective, rather than those finally found justified by the Commission, shippers 
and consumers would be paying almost $1 billion more in higher rates this year 
than they are now paying. Moreover, now that the cost figures have been gathered 
for the periods covered by the railroads' estimates in these proceedings it is 
clear that in their submissions, the railroads cumulatively over-estimated their 
costs by at least $1 billion annually. Obviously, if the Commission had not taken 
the time to examine and weed out these unjustified increases, the public would 
now be paying these unreasonably high rates.^^ 

On the next page, former Chairman Stafford continues: 

The contention is advanced that regulation is largely responsible for the plight 
of those railroads that are in economic diflaculty and that this bill will remove 
a major roadblock to profitability. We dispute the premise that regulation has 
had a substantial impact on the economic troubles of some railroads. AVe note that 
numerous factors — the unavailability of long-haul traflic in the Northeast and 
the relatively high cost of doing business in that region, the effects of the present 
major reces<^ion, the unequal subsidization of other modes of transportation, 
archaic work rules, management problems, et cetera — have contributed to the 
financial weakness of most of the railroads in the Northeast and some in the 

But we do not believe that there has been any persuasive showing made that 
regulation has substantially contributed to the plight of railroads in this region. 
On the contrary, the overall relative health and stability of the other modes of 
transportation regulated by the Commission and of the other railroads outside 
of the region suggest that the root causes of the problem are those factors specifi- 
cally applicable to the weak Northeastern and Midwestern railroads, rather than 
across-the-board Commission regulatory policies." 

The contrast between these two quotes requires no further emphasis. 
Perhaps it might be pointed out, however, that Eastern District rail- 
roads accounted for 89.2 percent of all rail freisjlit tonnage, and 83.6 
percent of all rail freio'ht revenues, in 1974.^^ But it is clear that the 
Interstate Commerce Commission claims responsibility for massive 
savings in rail freight rates, while taking no responsibility for such 
presumed negativ^^ factors as ''archaic work rules" and "manasfement 
problems." In a highly labor-intensive industry,^® it is hard to see how 

51 Ibid., p. 73. 

52 Transport Economics, vol. II. No. 2 (1975), p. 2. 

53 "T'.ailroad RevitaUzation," Hearings . . . House of Representatives (op, cit. ), 1975, 
p. 202. 

s*Thid., p. 203. 

55 Calculated from "Transport Economics," vol, II, No. 2 (1975), p. 8. 

58 In spite of the seeming paradox of describing the railroad industry as both "labor- 
Intensive" and "capital-intensive," both descriptions are correct. The explanation is two- 
fold : fl) Since railroad enrnintrs per unit of capital omnloyed are exceptionally low, 
Pleasuring "capital-intensiveness" in the usual way, by the ratio of capital to revenues, 
greatly over-states the relative importance of profits ; (2) Since railroads have an unusually 
high ratio of "value added" (direct payments to employees, owners, and creditors of the 
"business, etc, as opposed to external payments to suppliers of fuel and raw materials), 
to operating revenues, wages — which are the most direct measure of "labor-intensiveness" — 
are a relatively high proportion of revenues even though they are relatively not so 
important as a proportion of value added. 


it is possible to achieve cumulative saviiiirs of Si billion per year, com- 
pared to amiual freio-ht revenues of some $16 to $17 l3illion, in the 
face of the continuation of "archaic work rules" which are certainly 
not peculiar to either the Northeast or the Midwest. Hence we are led 
to a mystery : Is the Interstate Commerce Commission responsible for 
the kinds of savings mentioned by former Chairman Stafford? If so, 
how is it done ? And what may this portend for the future ? 

The first key toward unlockins: the mystery is contained in the dis- 
tinction between an index of freiofht rates and of revenue per ton-mile. 
Former Commissioner AValrath's statistics for 1951-69 must certainly 
have referred to "railroad revenue per ton mile" — not, as he described 
the conc<?pt, "Carload freight rates." The former, unlike the latter, is 
affected by commodity mix, averasre lenofth of haul, unit of service 
(e.:^.. unit trains vs. carloads), etc. Revenue per ton-mile has been kept 
low by the loss of hiirh-rated freiirht traiTic to the truckers. 

The second key is contained in the further question of how to meas- 
ure the productivity of railroad labor. If taken at face value, the 
"Walrath data must either indicate a massive decline in profita- 
bility of the industry over the period covered by his statistics ^"^ 
or a massive increase in labor productivity. The increase in railroad 
waire rates of 111 percent was exactly equal to the increase in compen- 
sation per hour in private non-farm employment over the same period ; 
yet private non-farm output per hour went up by only 51.6 percent 
between 1951 and lOGO. Tliese differential rates of increase must have 
been the most important sinirle factor in the increase in the implicit 
Gross National Product price deflator for private non-farm output 
of 49.2 percent between 1951 and 1969. ^^ So what Ave are still faced with 
on the basis of the AValrath evidence is an identity in wacfe rate in- 
creases and a very substantial difference in price increases (9.8 percent 
for railroad "rates," vs. 49.2 percent for the implicit price deflator). 

The explanation for the apparent inconsistency probably lies in 
two factors. The first is related to railroad loss of traffic. This has been 
differential, and larofely concentrated on the more costly types of 
shipment. It has been concentrated on the shorter hauls — averacre 
railroad haul per ton increased from 419.99 miles in 1951 to 496.82 
niiles in 1969, or by 18.3 percent ^^ — and it has also been concentrated, 
bimodally, on manufactures and miscellaneous })roducts which moved 
over to motor carriers, as well as on certain bulk products accessible 
to waterways which moved over to barge, or on coal movements which 
were replaced by oil and natural gas. 

By far the most important and pervasive of these competitive in- 
fluences was that of the motor carrier. Freight revenues of Class I 
and II line-haul railroads dropped by 5.9 percent between 1955 and 
1960. and increased by 36.5 percent between 1960 and 1970 for a total 
percentage gain of only 28.4 percent over the entire 15-year period. 
The comparable percentages for Class I, II. and III motor carriers of 
property were: +30.3 percent between 1955 and 1960; +102.2 per- 
cent between 1960 and 1970; and +163.5 percent, 1955-1970.^° This 

*r>"/.^^l^j^^ *^ ^^^^ extent in fact occurred : Net operating Income of railroads was 
Sn.)r,.no9.000 m lOol and $GG7.1.->7.000 in I'JGO, Historical Statistics of the United States, 
Part 2. p. 730. 

68 Private non-farm data from Economic Report of the President, Januarv 1976, Table 
B-SO. p. 20R. 

59 Historical Abstract. Part 2, p. 732. 

<» Transport Economics, vol. II, No. 1 (1975), p. 19. 


compares with a growth in money Gross National Product of 146.0 
percent over the same period.^^ So the regulated motor carrier industry 
was squarely in the mainstream of American economic growth. It 
managed to attain this position wdiile the railroad industry was very 
conspicuously lagging. 

Therefore ': given the measurement of the productivity of railroad 
labor in ton-miles, increase in average length of haul would, by itself, 
increase reported productivity. So would loss of traffic in manufac- 
tured goods which previously involved the necessity of various kinds 
of expensive premium service — and the compensating luxury of a 
premium rate. 

A second obvious source of apparent railroad productivity gains 
since 1951 has been exploitation of the possibility, over considerable 
periods of time, of deferring maintenance. This may be a genuine 
source of gains in situations where a branch line is eventually to be 
abandoned, in any case; but, as revelations about northeastern rail- 
roads in recent years have made amply clear, much of the deferred 
maintenance was simply under-maintenance. Railroad maintenance is, 
in effect, considered as an operating expense. So deferral of mainte- 
nance not only put off the evil day in a physical sense ; it also put it 
off in a financial sense. Thus the same short-run decision will both 
serve to enhance apparent labor productivity in at least the earliest 
period of deferred maintenance, and enlarge current reported profits 
in this same period. But unless the line is to be abandoned, this kind 
of gain in "productivity" is simply a device of shifting labor costs 
from the present into the future. 

Therefore the billion-dollar "saving" of former Chairman Stafford 
must certainly be confronted w^ith such questions as: Is this saving 
cumulative for the longer-run, or is it bound to reverse itself with com- 
pound interest at some later date ? Is it bought by lowering of actual 
costs, or by purely accounting excisions of purely accounting costs? 
And, to the extent that the "saving" actually causes alterations in the 
day-to-day operations of railroads, will they gain more business by the 
putatively lower rates they are charging, or lose more due to inferior 
service ? 

Finally, there is the question of how much of this cumulative $1 
billion was a purely paper figure from the outset. Railroads, like all 
other claimants, are not disposed to understate either their case or 
their suggested relief before regulatory bodies. Railroads, unlike some 
other claimants, often use general rate increases as the occasion for 
a certain amount of reshuffling of rate structures through "holddowns" 
(the actual rate increase is held below the permitted increase, usually 
by expressing the increase in dollars instead of percentages) , and "flag- 
outs" (no actual rate increase, in spite of blanket rate increase per- 

So two questions remain: how much of the $1 billion is substance 
and not shadow? And, to the extent that it is substantive, what 
are its implications for tomorrow ? Behind both of these specific ques- 
tions lurks the general question : to what extent is regulation of rate 
levels of special economic importance in periods of general inflation ? 

« Economic Report of the President, January 1976, Table B-1, p. 171. 


Regulation as a means of retaining desired relationships heticeen dif- 
ferent rates charged hy different modes^ or different rates charged 
hy the same mode 

In this discussion, Tre must begin by facing a paradox : As was em- 
phasized in the historical section of this paper, the social forces that 
gave birth to the Interstate Commerce Commission were apparently 
more opposed to rate discrimination by railroads than to all other 
rairoad evils put together. Beginning in 1920, the Interstate Com- 
merce Commission has had legal power over every aspect of railroad 
rates, and beginning in 1935 over every aspect of interstate motor 
common carrier rates on non-exempt commodities ; the general public 
impression still is that "the Interstate Commerce Commission sets 
freight rates"; yet, according to tlie Commission's own submissions: 
it suspends very few new rate filings; it subsequently allows most of 
these suspended rates to go into effect : and. all in all, its influence is 
confined to a mere fraction of 1 percent of all proposed rate clianges. 

This paradox cannot be resolved either by appeal to numbers of 
by a neat application of economic logic. So let us pursue the topic 
numerically as far as that is possible, and then continue into the 
thickets of opinion. 

Class rates^ where the ICC has initiated major ex parte action on rate 
structure^ are unimportant 
The one area in which the Interstate Commerce Commission has 
conducted massive investigations, on its own motion, is that of class 
rates.^- This has also been the area in which systematic rail cost infor- 
mation was first compiled — in connection with Docket Xo. 28300, 
"Class Rate Investigation,'" 1939.^^ But, even as the I.C.C.'s first sys- 
tematic effort to relate rates to costs was under way, class rates them- 
selves had lost practically all their impoi-tance. Of the rail carload 
traffic originated Seyjtember 23, 1942, only 4.1 percent of the carloads 
moved on regular classification ratings, while 10.7 percent moved on 
exceptions rating and the remainder on commodity rates.^* Although 
exception rates were tied in systematic ways to class rates, their aver- 
age reduction ranged from 20.7 percent in Official Territory (the 
northeast) to 24.5 percent in Southern Territory .^^ As for commodity 
rates — they followed, and follow, no single rule. Some, designated as 
"column rates.'' reflect uniform percentage reductions from the class 
rate which would otherwise apply ; others may be systematically based 

"2 "Class rates*' may be defined as the hooks on which the entire railroad rate structure 
depends. Consequent upon the Interstate Commerce Commission's "Class Rate Investiga- 
tion." 1939, the railroads managed to work out a uniform classification of all freight which 
was made effective on May 30, 1952 ; the Commission's 1945 decision in the Class Rate 
case had already established uniform class rates east of the Rockies. The Commission's 
decision established thirty different rate classes to wliich one was subsequentlv added for 
a total of thirty-one. The highest was Class 400; the lowest was Class 13 (i.e.' 13 percent 
of Class 100). The ratio between the highest and the lowest rate was therefore 400/13, 
or 30.77 :1 (see Locklin. Economics of Transportation, pp. 175-176). 

^ This was donp under the supervision of Ford K. Edwards, who had previously developed 
his principles of rail costing in California. The first general attempt at determining cost 
differences — which, of course, implied work on cost levels — was brought out bv the I.C.C. 
Bureau of Statistics, undpr the supervision of M. O. Lorenz. in 1930 ("Territorial Variation 
in tht Cost of Carload Freight Service on Steam Railways In the United States for the 
Tear 192S"— Statement Xo. 3018). 

«* Board of Investigation and Research, Report on Interterritorial Freight Rates f78th 
Consress, 1st Session, House Doc. 303 — Sept. 27, 1943), G.P.O., Washington, D.C., 1943, 
p. 48. 

^ Ibid., Table 30, p. 48. 


on different principles; still others consist of special point-to-point 
rates with no systematic milea^2:e or other basis.^^ 

In revenue terms, the comparison was onl}^ slio^htly more favorable to 
class rates. On September 23, 1942, they provided 6.3 percent of car- 
load revenues as against 16.1 percent for exception rates and 77.6 per- 
cent for 'commodity rates." Althoujjh more recent data are not avail- 
able in comparable form, probabilities are that rail movements under 
class rates have declined even from their low level of 1942.^^ ^ 

Therefore it is temptins: to say that the one segment of the rail rate 
structure with respect to which the Interstate Commerce Commission 
has shown sio-nificant initiative— the class rate structure— is both the 
least important of all seg^ments and also the least likely to govern fu- 
ture traffic movements. This conclusion would be rather unfair, since 
numerous other rates are keyed to class rates, and since particular com- 
modity rates or rate structures on important bulk commodities — such 
as coal— are bound to be brought to the attention of the Interstate 
Commerce Commission without any action on the Commission's part. 
It is nevertheless true that, with respect to rate revisions, the Com- 
mission's role deserves the description "quasi-judicial." The Com- 
mission is essentially an arbitrator or referee, not a molder of the 
economics of the transportation system. 

Rate hureaus make an important contribution to the establishment of 
specijiG rates 

This brings us to a controversial subject : if the Commission is a 
referee, who has already set the rules of the game ? Specifically, how 
much genuine independence do individual firms possess in the field 
of transportation, and to what extent do rate bureaus impose cartel 
restrictions on individual firms ? 

Eate bureaus have always subsisted in an economic twilight zone. 
In the bleak first decade of the Sherman Act, about the only victories 
the Government was able to win were the consequences of Section 1 
cases against railroad rate bureaus.^^ Thereafter rate bureaus in both 
rail and motor transportation came under sporadic attack, until they 
were protected by the Reed-Bulwinkle Act of 1948.'^° Even this legis- 
lation contains a saving clause limiting joint intermodal carrier activ- 
ity to "matters relating to transportation under joint rates or over 
through routes." ^^ Moreover, whether for legal or practical reasons, the 
rate bureau strongholds are strictly within each mode. But the legal 
position of the rate bureau has seemed relatively secure since 1948. 
What has been its economic position? 

Nominally a "rate bureau" is simply the instrumentality employed 
for joint publication of rates. Given an institutional structure which 

63 Ibid., p. 92. 

«^ Milton S. Heath, "The Uniform Class Rate Decision and Its Implications for Southern 
Economic Development," Southern Economic Journal, vol. 12 (January 1946), p. 218, 

68 As might be expected, motor carrier rates do not cover as wide a range, and probably 
do not show as much irregularity, as rail rates. The lowest motor carrier rate classifica- 
tion is significantly above the lowest rail classification in all parts of the country ; and 
I.C.C.-regulated motor common carriers bill a much higher percentage of their business 
under class rates. This is practically inevitable given the importance of lesp-than-truckload 
traffic to regulated common carriers. But, even in the motor carrier industry, commodity 
rates are both tailor-made and important. 

69 United States v. Trans-Missouri Freight Association, 166 U.S. 290, and United States v. 
Joint Traffic Association, 171 U.S. 505. 

TO Interstate Commerce Act. Part I. Section 5a. 
^Section 5a (B) (4). 


relies to a considerable extent on traffic interchange, and a legal sys- 
tem which relies on fixed as well as published rates, an arrp.ngement 
of the rate bureau type is practically inevitable. But publication im- 
plies both notification of rate changes, by carrier-members to the rate 
bureau, and filing of the proposed new rate with the Interstate Com- 
merce Commission. Hence the almost necessar}^ development of the 
"open pricing" aspects of cartel organizational structures. Public fil- 
ing would produce some of these aspects even without rate bureaus, 
especially since rate changes must be filed in advance of their appli- 
cability. But operating through rate bureaus clearly enhances possi- 
bilities for pressure by competitors (or potential connections). 

Here there are distinctions between railroads and motor carriers. 

The number of rail competitors on given routes has continued a de- 
cline which began soon after "World TVar II. Xot all of the railroad 
mergers of recent years have combined parallel railroads, but there 
has been enough reduction of competition to reduce the prospect for 
many possible alternatives except on very long hauls or between a few 
very important railroad centers. On the other hand, the motor carrier 
industry has alwaj^s had more competitors on major routes or between 
major centers, and doubtless still does. To the extent that only one or 
two firms serve a market, as is often true for railroads, rate bureau in- 
fluence on single line shipments is diminishing to the point of disap- 
pearance. Converseh% rate bureaus would be expected to have more 
independent importance in the trucking industry. 

For technical, economic, and historical reasons, unit railroad costs 
tend to diminish as traffic over a given line increases. But once motor 
carrier density goes beyond the load effectively permitted for one 
tractor-trailer combination by Federal and State weight and size limi- 
tations, there is no reason why costs of truckload movements should 
diminish, per truckload, with number of truckloads over a given 
stretch of hiirhway. Hence the incentive to cut rates is less for motor 
carriers, if the sole effect of the reduced rate is a doubled traffic vol- 
ume at approximately a doubled cost. 

Railroads have had rate bureaus, or their equivalent, for far longer 
than motor carriers. For rate bureaus or any similar co-operative en- 
terprise, age presumably helps to confer stability. 

Kailroad rate bureaus make a great deal more geographical sense 
than motor carrier bureaus. The averaire rail lenirth of haul is about 
75 per cent more than the average truck length of haul. Therefore one 
would expect each railroad rate bureau to cover a wider geographical 
expanse. But, after due allowance for this difference, there can be no 
explanation for the present multitude of motor carrier rate bureaus — 
including "leap-frogging" rate bureaus, which cover long-distance 
shipments originating in the territory of one rate bureau and termi- 
nating in the territory of another — except the historical fact that the 
motor carrier originated with essentiallv local haulage. To the extent 
that motor carrier rate bureaus are needlessly localized and needlessly 
numerous, their ability to influence through rates independently is 
possibly impaired. 

This impairment must be exacerbated by the I.C.C.'s inability to con- 
trol either through routes or joint rates in the motor carrier industry. 

The Interstate Commerce Commission has what might be called 
"control of the last resort" over intrastate rail rates. Intrastate motor 


carrier rates, which are relatively more important to the motor carrier 
industry, are beyond I.C.C. jurisdiction. Hence there is a special local 
role for rate bureaus in motor carrier transportation which does not 
exist for railroads. 

These differences do not all point in the same direction. But it ap- 
pears to be a fair conclusion that if motor carrier rate bureaus are to 
emulate their older rail counterparts, they must exert greater con- 
scious effort. And it would appear, from a statement made as early as 
1943, that the motor carrier rate bureaus were aware of this fact al- 
most from the outset : 

... In the judgment of the Board the underlying cause of oppressive tactics 
by motor-carrier rate bureaus is the apparent laclv of control of such organiza- 
tions by the member motor carriers. . . . 

A motor-carrier bureau . . . should not take action against the rates of its 
own member carriers by requesting suspension of such rates filed with the In- 
terstate Commerce Commission. Such practice lends itself too readily to restraint 
upon the right of the individual to name his own individual rates. . . .'^' 

Obviously, the mere existence of a rate bureau provides a forum for 
joint consultation, and a background for joint action. The rate bureau 
problem still is what it always has been : how to facilitate necessary 
rate co-operation without encouraging collusion? 

In both the railroad and the motor carrier industries, rate bureaus 
play a substantial role in the process of rate formation. For reasons 
to be examined in the next section, much of this role would exist even 
in the absence of anyone's conscious desire for agreement on prices, 
simply because so many transport transactions involve collaboration 
on ]^rices in the form of joint rates. But joint rates cannot be set in an 
institutional vacuum; and the institutional environment is certainly 
one which favors rate harmony among competing firms. Thus the fact 
that the Interstate Commerce Commission so rarely has to act as a rate 
policeman may be simply because the domestic quarrels on its beat 
have mainly been resolved without the necessity for outside inter- 

TJie rate-making role of the Interstate Commerce Commission 

^Vhat, then, remains for the I.C.C. ? 

It does, in fact, operate as the arbitrator (or policeman) of last 

Even though tlie I.C.C. suspends and investigates only a trivial per- 
centage of rate changes which are filed with it each year, this percent- 
age still amounts, in absolute terms, to hundreds of cases. Over the 
decades, decisions reached in hundreds of cases per annum cumulate 
into a formidable body of what amounts to case law. 

The Interstate Commerce Commission's consistency, from decision 
to decision, is not perfect but is considerable. If potential litigants 
already know the ansAver will be negative, there is no point in raising 
the question by proposing a rate change. 

Investigation and Suspension cases tend to be big cases, either in 
money value or in principle. Therefore the practical impact of I.C.C. 
decisions should be measured by a weighted aA^erage which would 
yield a figure far above the oft-cited fraction of one per cent. 

'^ Board of Tnvestlffatlnn and Research. Report on Rate-Makinj? and Rate-Publi^hlnj? 
Procedures of Railroad, IMotor, and Water Carriers (78th Conjrress, 1st Session, House 
Document ^o. 363— Nov. 29, 1943), G.P.O.. Washington, D.C., 1944 pp. 67-6S 


In short, recent Interstate Commerce Commissioners have been 
modest, indeed, in their citations of the Interstate Commerce Commis- 
sion's numerical partici]Dation in ths process of rate formation. The 
Commission has been highl}' selective ; it has been mainly passive ; but 
it has not been impotent. 

The nature of Interstate Commerce Conunission influence on rate 

Xow that vre have concluded that the Commission does indeed in- 
fluence rate structures, how is this influence exerted ? 

With respect to railroad rates. I. L. Sharfman's observation of the 
mid-1930's still provides a reasonably up-to-date summary. 

. . . Xo distinction is drawn between the absolute and relative reasonable- 
ness of particular rates, and prime reliance is generally placed upon the method 
of comparison with other rates.^ 

There were once two powerful influences which led to this result : 
relative rates were what most interested shippers (with special refer- 
ence to geographical and commodity rate relationships) ; and no one 
had any meaningful cost data on wliich to erect any other kind of rate 
structure. The second influence is less powerful than it was ; but, given 
the inherent structure of railroad costs, it will probably have some 
relevance as long as tlie industiT remains. The flrst is still very impor- 
tant in folklore, and has not lost its importance in fact. 

With respect to motor carrier rates, the problem is complicated by 
an attempt to impose a value-of-service Iron Maiden on an industry 
which is not only potentially but in considerable measui'o actually 
competitive. Possibly the most extensive study of Commission policy 
with respect to rail, motor, and intermodal rates has these comments 
to make about motor carriers : 

The ursre to establish rates upon this basis [i.e., the "added-traffic theory**] 
arises, usually from lack of balance in the traffic of a particular carrier and the 
accompanying desire to fill trucks which must otherwise move empty in order 
to balance equipment. . . . The limitations upon individual carriers in their 
route and commodity authority, their small size and restricted traffic connec- 
tions, and the varied success of their solicitation injure that all carriers will not 
be affected alike and that they will not necessarily reflect the a2r.2:reg:ate pattern 
in their individual operations. Moreover, the average motor carrier is a small- 
scale enterprise and back-haul pricing can reverse the jiredominant direction of 
its traffic movement more readily than that of a railroad.'* 

This description could not be improved on as a do|)iction of the di- 
lemma in store for regulation of an industry in which competition 
keeps threatening to l)i'eak out. Xor can it be bettered as a description 
of how such attempted regulation may create new economic problems 
in an attempt to do something about what are believed to be old ones. 
With respect to intermodal rates, the regulatory situation was 
changed significantly — but not by any means revolutionized — by the 
Transportation Act of 1958. Professor Williams- study, which ap- 
peared in that year.'-^ conveys an impression of umbrella rate-making 
a? a standard ])ractice. with motor carriers adopting railroad ]-ates 
whenever that seemed profitable, and refusing to accept traffic in low- 

"^ I. L. Sharfnian. "The Interstate Commerce Commission." part 3. vol. B, the Common- 
wen ith Fun(i. New Vork. 19.^,6. n. 10. 

'* Ernest W. William*. Jr., "The Regulation of Rail-Motor Rate Competition." Harper & 
Brothers. Xfw York. !;>.'.>. pp. l.').",. i;is-139. 

"Op. cit. 


rated categories. Profes^^or George Hilton's 1069 study of "The Trafig"- 
portation Act of 1958," ^^ duly considers the notorious Ingot Molds 
case ^^ and ends a review of a decade which was supposed to be devoted 
in part to reducing umbrella rate-making with the comment, ". . . one 
can still say . . . that the inherent ambiguity in the rule of rate- 
making remains unresolved." ''^ 

The extreme form of the intermodal rate umbrella, in spite of the 
Act of 1958, remains the Ingot Molds case. The misunderstanding of 
economic reasoning which was involved in the settlement of this case 
Jnust be laid at the door of the Supreme Court as well as the Inter- 
State Commerce Commission; but, although the blame can be dis- 
tributed, there is enough blame to go around. The issue involved in 
this case was of classic simplicity: railroads transporting ingot molds 
from Neville Island and Pittsburgh, Pennsylvania to Steelton, Ken- 
tucky wanted to establish a new rate which was, by I.C.C. costing 
standards, above the nearest available equivalent to railroad marginal 
or incremental cost of the service, about equal to the current barge 
rate down the Ohio River, but below the fully-distributed costs of 
competitive barge transporters. The Commission argued — according 
to the Supreme Court, successfully — that railroads should not be al- 
lowed to charge on a marginal cost basis if the resultant price was 
below the average cost of a regulated competitor. Quite apart from 
the question of the practical extent and significance of "regulation" of 
barge rates, this decision flew in the face of professional analysis of 
railroad economics extending back to the 1850s and even the 1840s. 

The Ingot Molds case cast a shadow over a feature of railroad rate- 
making which has been recognized by railroad men, economists, and 
regulatory bodies practically from the origins of the industry. This 
feature is the employment by the industry of "value of service" rates. 
Also described by hostile critics as "charging what the traffic will 
bear," these rates are the only method available to a firm whose unit 
costs decrease with greater density of business (longer trains; more 
trains per mile of track; in general, more revenue tonnage per mile 
of line) for the simultaneous accomplishment of two desirable, but 
not wholly compatible objectives: (1) coverage of total costs; and (2) 
expansion of density of use of facilities so that no customer from 
wliom any net return can be derived is turned away. 

From the first appearance of the railroad industry, one of the prime 
determinants of "vahie of service" was the rate charged by the most 
nearly competitive form of transportation, whether this closest com- 
petitor was another railroad or another transport mode. A dilemma 
which had already appeared before 1887, ancl continues in perhaps 
exaggerated form today, is this : if all railroads set rates to undercut 
the marginal costs of the most closely competitive railroad, is there 
any assurance that they can find the revenue to cover the spread 
between their own marginal costs and their higher average costs? 
In short, in a freely-competitive railroad rate-making environment 
within w^hich railroads competed in fact as well as in theory, what 
would railroads find to live on? Hence one element in the continued 

"'^ Indiana University Press, Bloomington, Ind., 1969. 

7^323 I.C.C. 758 (1965) ; 326 I.C.C. 77 (1965) ; 268 F. Supp. 71 (1967) ; 392 U.S. 571 

78 Op. cit., p. 78. 


support for railroad rate bureaus and continued I.C.C. control over 
railroad minimum rates. 

But, if the closest railroad competitor is another mode of transporta- 
tion, and this mode is not characterized by decreasing costs in the sense 
in which this term is now being used, is there any economic significance 
to be attached by a regulatory body to the costs — average, marginal, 
variable, or whatever — of this competitor from another mode? The 
answer has to be, "No." The idea of "predatory competition" which 
developed, justifiably, in the early days of railroading is irrelevant to 
competitive railroad rates directed at competitors from other modes 
who are expanding, mobile, and not themselves capable of reducing 
their unit costs by increasing the density of their business. 

Therefore, in Ingot Molds, both the Interstate Commerce Commis- 
sion and the Supreme Court betrayed complete misunderstanding of 
economics in four respects: (1) They misunderstood the fundamental 
economic distinction between "average" and "marginal"; (2) They 
misunderstood the importance of "opportunity costs" (costs of the 
closest competitor) to the "value of service" element in railroad rate- 
making; (3) They misunderstood the distinction between unfair or 
predatory competition within an industry whose unit costs decrease 
with density, and between this industry and other industries with com- 
pletely different cost structures; (4) They overlooked the impossibility 
of practicing so-called "predatory competition" by differential rate 
cuts against firms all of whose assets (and costs) permit a high degree 
of geographical and functional mobility. 


The consuming interest in rate structure which was the main in- 
fluence behind both the Cullom Report and the Interstate Commerce 
Act has not disappeared. Possible geographical discrimination in reg- 
ulated transportation rates is still a very touchy issue. 

The central role of the Interstate Commerce Commission with re- 
spect to rate structure remains intact, even though the Commission 
chooses to suspend only a minute fraction of all proposed rate changes. 

But the center of gravity has markedly shifted. The case for regu- 
lation of rate relationships which was developed in the years before 
1887 was primarily geographical. It was also based on what some 
critics believed to be unreasonable commodity price discrimination 
and on a recent history of rapid railroad rate fiuctuations. Today, 
however, the motor carrier is practically ubiquitous. It provides trans- 
port services in relatively small lots. An unregulated motor carrier 
industry would be a particularly poor industry in which to attempt 
to develop a transportation monopoly, and a particularly poor instru- 
ment for fastening monopoly on other industries through extreme 
price discrimination. Therefore the main explanation for the often- 
exercised power of the Interstate Commerce Commission to determine 
rate relationships for the motor carrier industry seems to be either 
nostalgia or a misplaced analogy of motor with' rail transportation. 

IVFeanwhile, regulatory controfof rate relationships which was orig- 
inally designed to protect the relative position of individual shippers 
or communities had swung, in cases such as the Ingot Molds case, 
toward an attempt to use rate structure as a means of protecting spe- 
cific transport modes. This attempt was apparently a product both of 


a misunderstanding of the economic basis for "value of service" rate- 
making in the railroad industry, and of a misunderstanding of the dif- 
ference between the economics of railroading and the economics of 
motor or water carriage. However, the Ingot Molds case was an ex- 
treme example; and several provisions of the 4-R Act of 1976 may 
prove, when duly interpreted, to have combined to outweigh Ingot 
Molds. Section 202 of the 4-R Act contains several provisions (with 
respect to seasonal, peak, and regional rates, separate rates for distinct 
rail services, selective rate changes, for two years, within a 7 percent 
annual limit, etc.), whose final effects on railroad rate policy will 
probably be determined by eventual judicial determination of stand- 
ards for rail "market dominance." '^^ 

Regulation as a means of controlling the institutional structure of the 

(a) By entry cont7'ol 

Although Interstate Commerce Commission control of entr}^ is 
legally just as strong for railroads as for motor carriers, the railroad 
control is without important practical consequences. Quite apart from 
the generally discouraging financial and economic record of the rail- 
road industry since 1929 (barring World War II), successful entry 
into railroading could not be accomplished with a single capital unit as 
small as one motor truck. 

But control of entry in trucking is very important. This is true not 
only because of the fact of control, but because of the type of control. 

First, control serves to subdivide the industry into practically non- 
competing groups : e.g., of regular-route common carriers of general 
commodities, and of irregular-route carriers (usually of one or a few 
specific commodities, or commodities requiring specialized transporta- 
tion equipment). 

Second, it helps to determine the relative profitability of individual 
carriers within each group. Operating rights can be a valuable intangi- 
ble asset. In terms of ability to apply for operating rights all actual 
and potential motor carriers are equal; in fact existing carriers are 
much more likely to be successful in getting new authority than pro- 
spective new^ carriers. 

Third, for less-than-truckload shipments which require terminal 
and trans-shipment facilities and are accompanied by shipper distaste 
for any unnecessary transfers and liking for single-carrier routing 
whenever possible, the extensiveness of regular-route operating rights 
is in itself an important asset. Less-than-carload shipments involve 
systems effects. A larger system, in terjus of mileage of operating 
rights, can attract business away from a smaller or more localized sys- 
tem. Therefore both the total attractiveness of less-than-truckload 
service rendered, and the differential attractiveness of how that service 
is rendered by different regular-route motor common carriers of gen- 
eral commodities, are influenced by Interstate Commerce Commission 
entry decisions. 

• W ^Pl^'ss the Commission finds market dominance to exist under the Seven Per- 
centuni Provision or the likelihood of its existence under normal procedures the Commis- 
sion has no power to suspend any rate on the ground that it may be unjustly or unrea- 
sonably hijrh . . . ' (Interstate Commerce CoTumission, "The Impact of the 4-R Art Rail- 
road Rule-makins Provisions," October 5, 1977. p. 1). ^^ "'^^ 


The intermodal effects of this control of entry into the motor carrier 
industry are less important than the intramodal effects. A rule-of- 
thumb for the motor carrier industry is : the larger the shipment, the 
smaller the firm that can perform the transport operation most effi- 
ciently. Owner-operators are helpless in face of the permutations and 
combinations involved in the operation of any extensive less-than- 
trucldoad business. Very large firms would have difficulty in com- 
peting on manv full truckload shipments with an owner-operator. An- 
other rule-of-t'humb for the motor carrier industry is : the smaller the 
shipment, the more likely it is to be transported by a regular-route 
common carrier. Irregular route carriers extend right up to the fron- 
tier of contract carriage. The only differences may be legal, in tliat they 
have more than seven clients, or economic, in that they are prepared 
to submit to rate control beyond control of minimum rates. Regular 
route common carriers are typically, with respect to the bulk of their 
business, a long way from emphasis on heavy movements of truck-load 
lots, and therefore a long way from contract carriage, a long wa}^ from 
what railroads do best, and a long way from head-to-head competition 
with railroads. 

Finally, control of entry occupies a high proportion of the Inter- 
state Commerce Commission's time. ^lotor carrier operating authority 
cases closed by the Commission during fiscal 1976 amounted to 6,800 
out of 8,857 cases — or 76.8 per cent of the total. "^ 

(b) By control of mergers 
Interstate Commerce Commission jurisdiction over mergers is not 
"regulation" in the sense that the word is used in most of the rest of 
this discussion. Section 5(11) of the Interstate Commerce Act says, 
among other things, 

The authority conferred hy this section shall be exclusive and plenary, and 
any carrier or corporation participating in or resulting: from any transaction 
approved by the Commission thereunder, shall have full power ... to carry 
such transaction into efifect . . . and any carriers or other corporations, and 
their ofl5cers and employees and any other persons, participating in a transac- 
tion approved or authorized under the provisions of this section shall be . . . 
relieved from the operating of the antitrust laws and of all other restraints, 
limitations, and prohibitions of law, Federal, State, or municipal. . . .^ 

The locus of control over mergers is especially important for rail- 
roads. The term "Balkanization" has already been employed in this 
paper, and it will receive heavy emphasis later on. But its possible 
ill effects on the railroad industry are so obvious that they can be 
summarized briefly : inappropriate investment policies with respect 
to freight cars and inefficient use of them : inability to match the track- 
maintenance standards required by particular pieces of equipment 
with actual maintenance throughout the country; inability to mount 
a unified marketing effort in view of an inability to guarantee service 
standards insofar as they involve "foreign" railroads; internecine 

■^ statistics in this paragraph are derived from I.C.C, "90th Annual Report to Congress" 
1976. App. B. Table 1. p. 114. 

^The Interstate roninierce Act (as amended to 1073). op. cit.. pp. .3.^—^4. Section 401 
of the 4-R Act of 1076 confers additional powers with respect to mergers and merger 
planning on the Secretary of Transportation, but still within the frnniework of ultimate 
I.C.C. jurisdiction. For the Conimissioirs most recent interpretation of this changed situa- 
tion, see I.C.C. Rail Services Planning Office. -'Rail Merger St\id.v : rrelinilnary Report." 
November 1, 1077. The Office recommenf^s. inter nlin. that the CommissioTi issue a policv 
statement "Endorsing end-to-end mergers." "Empliasi/ius: . . . voluntary agreement 
among carriers . . .". and "Declaring that mergers are not a proper and effective method 
for dealing with marginal carriers . . ." (op cit., p. 2). 

S3-944— 78 4 


warfare over rate divisions, car-hire per cliems, and so on. The "Bal- 
kanization" list of problems is a very long list. 
But a mere recital of the list does not answer three key questions : 

( 1 ) even assuming that the railroad industry is still too fragmented, 
in terms of ownership, to be fully efficient,^^ can anyone seriously 
believe that indiscriminate railroad merger activity would be per- 
mitted without any governmental oversight? 

(2) given this necessity for oversight, would a merger route which 
proceeded directly via Federal executive agencies and the courts be 
less time-consuming, more attuned to specific transport problems, and 
generally more economically efficient than present procedures ? 

(3) regardless of the governmental location of oversight, can "Bal- 
kanization" of the railroad industry be alleviated purely by the meth- 
ods developed by the Interstate Commerce Commission, or the De- 
partment of Justice and Federal Trade Commission? 

To sort out these three questions : Question One is meant to be rhe- 
torical. Question Three is rhetorical in the sense that the answer has 
to be "no." But it does not fade from the scene with a mere verbal re- 
buff : this "no" answer either throws the issue back to Question Two, 
or points toward some as yet unexplored possibility. And Question 
Two involves an important policy issue to which no one can give a 
definite answer. 

Tlie horrible example with respect to the Interstate Commerce Com- 
mission's interpretation of its merger mandate is the Rock Island case. 
The incumbent President of the Rock Island does not blame the Inter- 
state Commerce Commission for this reductio ad dhsurdum of the reg- 
ulatory process : 

. . . there is something radically wrong with the system of government in- 
volvement in railroad merger proposals. I cannot blame the Commission . . . 
when the law itself is such that Hamurabi, Solomon. Moses and Justinian to- 
gether could not possibly provide all the due process that is called for. It would 
be a sad mistake for the legislative branch of government to chastise and berate 
the Interstate Commerce Commission for the travesty that is the Rock Island 
merger case. Given the circumstances, I would say that the various members who 
have sat on the Commission since 1963 have done a creditable job simply by 
issuing a report a year ago [i.e., in 1974] tentatively recommending the Rock 
Island merger.^ 

But then he adds a sad and evident truth : 

There is a problem now, however, Mr. Chairman. The Rock Island has been 
kept waiting at the altar for so long — 12 years — that we've accumulated a lot 
of wrinkles, and our suitors don't want us anymore. So we're going it alone, 
not making the assumption that the Midwest rail problem can be solved through 
merger and acquisition. It takes too long. . . .^ 

In the context of the present discussion, the significant features of 
what happened to the Rock Island are (1) the twelve years of elapsed 
time from start to finish of the case; (2) the fact that, by the end of 

81 And even this may be in some respects a gratuitous assumption. See Robert E. GaUa- 
more, "Railroad INIergers : Costs, Competition, and tlie Future Organization of the American 
Railroad Industry," unpublished Ph. D. dissertation, Harvard University, Cambridge, 
Mass.. May 1968, vphere the point is strongly made, and supported, that many railroad 
mergers of the post-World War II years were disappointments even before the Penn Central 

82 Testimony of John Ingram. In "Regulatory Reform — 1975." hearings before the Com- 
mittee on Government Operations, U.S. Senate. 94th Congress. 1st Session, Pursuant to 
S. Res. 71, G.P.O., Washington, 1976, p. 53. (Sections 202 and 203 of the 4-R Act of 
1976 were inserted in an effort to speed up the process of I.C.C. examination of merger 

83 Loc. cit. 


this time, a railroad which had been avidly sought by various con- 
tenders in 1963 was too "wrinkled" to be wanted by anyone; (3) the 
strong possibility that a rational disposition of the Rock Island would 
have required "a major restructuring of railroads in the West"; ®* and 
(4) the final fact that the Interstate Commerce Commission, when it 
had tools which might have been used for such restructuring during 
the 1920s, instead did everything within its power to avoid the task. 
Anyone who believes that the railroad ownership map of the United 
States requires major revision would be exceptionally naive if he ex- 
pected the Interstate Commerce Commission to be capable of it (even 
if it were legally empowered to do so). He would be equally naive to 
look toward the Federal Trade Commission or the Department of 
Justice — especially if his goal were the encouragement of mergers. 

It must be added in fairness that the Eock Island case is the extreme 
example. Critics who are disposed to see no good in the Interstate Com- 
merce Commission might counter that, on the contrary, the Penn Cen- 
tral merger case was processed too rapidly — if, indeed, the merger 
should have been approved at all. But, both in Penn Central and in 
Eock Island, as well as in all other merger cases, the Commission has 
been faced with the following statutory commandment : 

In passing upon any proposed transaction . . ., tlie Commission shall give 
weight to the following considerations, among others: (1) The effect of the pro- 
posed transaction upon adequate transportation service to the public; (2) the 
effect upon the public interest of the inclusion, or failure to include, other rail- 
roads in the territory involved in the proposed transaction; (3) the total tixed 
charges resulting from the proposed transaction; and (4) the interest of the 
carrier employees affected." 

To the extent that these issues are raised only in merger cases, 
then there is the obvious danger that mergers may either be delayed 
forever by "failure to include" other railroads, or loaded with cost- 
enhancing provisions as a consequence of some of the other stipula- 
tions of this paragraph. The basic question, therefore, becomes not 
just the efficiency of the Interstate Commerce Commission in carrying 
out its legal mandate with respect to mergers, but the appropriateness 
of this legal mandate (even as amended by the Act of 1976) to the 
present economic position of the American railroad industry. 

Begulation as a m-eans of controlling service standards 

{a) In general 

The Interstate Commerce Commission, along with other regulatory 
bodies, has legal powers not only with respect to price but also with 
respect to service. Through no fault of the Commission's, however, 
the service problem is a great deal more difficult for freight transporta- 
tion than for transportation in general, and more difficult for trans- 
portation in general than for the public utility industries strictly 

Take two examples from the utility field : electric power, and natu- 
ral gas. In both cases, the basic service attributes of the product are 
easily determined, easily controlled, and easily tested. There can be 

^* Jhid.. letter of former Chairman Stafford of the Interstate Commerce Commission to 
Sen. Abraham Ribicofif. Chairman. Committee on Government Operations, p. 2,'>7 (comment- 
ine on the February 1973 decision of the Administrative Law Judge in the Rock Island 

85 Interstate Commerce Act, Section 5 (2) (c). 


\'ery little imprecision, or argument, with respect to voltage, frequency^ 
the accuracy of meters, British thermal units per thousand cubic feet, 
etc. Passenger service at least approaches some of these "automatic'^ 
standards: trains and buses have schedules; trains and buses sup- 
posedly operating on these schedules can be expected to make a bona 
fide attempt so to operate ; and, once the operation has been completed^ 
variables' such as percentage of on-time performance can be quantified. 
The same is not true for freight. Freight trains may run on schedule^ 
but this can yield no assurance %vith respect to freight cars. Moreover, 
there would be no point in trying to enforce a higher standard of serv- 
ice than the shipper is interested in. He cares about overnight or sec- 
ond or third morning delivery, not about five minutes of lateness. 

Going beyond generalizations to specific modes of transportation^ 
the early history of I.C.C. regulation of railroad freight service first 
revealed an emphasis on equalization (in an attempt to minimize con- 
cealed discrimination) rather than on improvement. Today, it is obvi- 
ous that the condition of railroad way, stinicture, and equipment is one 
of the most important causal factors in determining the level of serv- 
ice. Slow orders and derailments inevitably mean bad service. Yet the 
I.C.C. has little leverage over maintenance policies. 

In the motor carrier industry, routes on which many carriers operate 
are likely to receive "standard" service with or without an attempt 
at direct regulatory intervention. Under such competitive conditions, 
a system of regulated minimum rates is likely to push all carriers 
toward uniform service whether individual shippers want this love! 
or not. Conversely, the small shipper in the small town is the last 
to obtain good motor carrier service, in the presence of regulation 
as well as in its absence. The only real sanction of the Interstate 
Commerce Commission in this case is its abilitv to grant new operat- 
ing rights over the same routes if it feels the incumbent is not doincr 
an adequate job. But since "inadequacy" is a subjective concept, how 
can witnesses served bv the deficient carrier oflfer any standard of 
comparison if this carrier has a monopoly? And how can they avoid 
his wrath if he retains it? And, most important, why would any 
Jiew entrant wish to break into a market which is not big enough to 
provide adequate profit for one firm ? 

(h) Railroad line ahandonment 
Tlie most hotly-debated area of service standards has been that of 
railroad proposals for line abandonment. The abandonment issue 
is important not only in itself, but also because it provides a classic 
example of what regulatory bodies were set up to do. what they were 
once able to do, and what they can no longer do with geographical 
effectiveness but without great long-run economic damage. 
This issue has already gone through two distinct stages: 
The first lasted from 1920 more or less up to the Penn Central bank- 
ruptcy. It involved essentially the same principles, and even the same 
lines of argument, as those which were traditionally used with respect 
to the maintenance of passenger service. The fii-st line of argument 
pertained to whether or not branch lines were, in fact, losing money — 
including an ambiguity as to whether the net revenue contribution of 
branch-line traffic to other lines should be counted in toto or only to 
the extent that it was retained by the specific railroad enterprise which 


owned the branch line. The second line of argument related to whether 
or not the particular branch-line traffic in question should be expectea 
to lose money. This raised the same issue of ''justifiable cross-subsidiza- 
tion*' that was typically raised, on a larger scale, by railroad provi- 
sion of passenger service. The issue that was carried down from the 
•days of "railroad monopoly," without being carefully re-studied by 
Congress or the Interstate Commerce Commission, was the issue o,f : 
if the railroads are to subsidize (or an individual railroad is to sub- 
sidize) branch line service, what are they to subsidize with? 

The second stage was foreshadowed as long ago as the 1920s, by a 
railroad which was already beginning to be something of a pioneer 
in misfortune — the Boston & iMaine. The Boston & Maine was the 
first important railroad to come foi'ward with a massive abandon- 
ment plan, which was turned down almost in its entirety by the Inter- 
state Commerce Commission on the railroad's first try. Later, little by 
little over the years, practically the entire original B & M plan was 
realized — but only after the railroad's vicissitudes were obviously 
such that drastic measures simply had to be taken. ^^ 

The Boston & Maine syndrome of the 192()'s eventually became the 
national railroad problem of the lOTO's: large and inescapable passen- 
ger defirits. and so forth, plus substantial losses on branch lines. These 
I)ranch lines may have had no independent economic justification, as 
transportation enterprises, in the first place. If the lines had had such 
a justification, as was true of some of the '"dowager" railroads which 
eventually became incorporated in the Boston & Elaine system, this 
had often been eroded by declines in freight tonnage, or in tlie unit 
Talue of this tonnage due to motor carrier competition, or both. 

So, by the late 1960's, the ad hoc api)roach used by the Interstate 
Commerce Commission with respect to abandonments.'and apparently 
acquiesced in by Congress, was plainly inadequate in view of the eco- 
nomic storm gathering in the Xorth'east and to some extent in the 

The first general answer to tliis storm, the Eegional Rail Reorgani- 
zation Act of 1973 (Public Law 9;]-2r>6), was instrumental in bringing 
into the open economic issues like traffic density, cross-subsidization, 
and low density mileage v. the kind of grudging step-by-step retreat in 
the face of an economically untenable position which was already fore- 
shadowed^by the Boston & Maine abandonment request. Moreover, the 
Act of 1971;^ broke new ground in stating a general affirmative public 
responsibility for support of large portions of the American railroad 
system — not just a responsibility for passenger service, which had al- 
ready been assumed by Congress under the terms of the Rail Passenger 
Service Act of 1970, and not just a negative responsibilitv of the type 
Avhich had historically been discharged through Congressional man- 
dates to the Interstate Conmierce Commission. Federal subsidies to 
radroads are as old as the land-grant era which opened with the orio-- 
mal construction of the Illinois Central. The idea was revived in^a 
modest way with the loan-guarantee provisions of the Transportation 

^^^'f °V° excellent detailed description of the Boston & Maine's frustrations with respect 
na?v^^??^°niv''rVrP^"^'^^-K^':'."^°^>^.°°' "^^^ Kegulatlon of Railroad AbandoSSs"' 
B ^M ''/7M^7f'' Cambridge Mass., 1948. The Cherlngton book also places the 
B & M episode within the context of I.C.C. policy toward abandonments in general. 


Act of 1958. But, in modern times, nothing has approached the sweep 
of Section 101(b) of the Eegional Kail Eeorganization Act: 

It is therefore declared to be the purpose of Congress in this Act to provide 
for — • 

(1) the identification of a rail service system in the midwest and north- 
east region which is adequate to meet the needs and service requirements 
of the region and of the national rail transportation system; 

(2) the reorganization of railroads in this region into an economically 
viable system capable of providing adequate and efficient rail service to the 
region ; 


(5) assistance to States and local and regional transportation authorities 
for continuation of local rail services threatened with cessation; and 

(6) necessary Federal financial assistance at the lowest possible cost to 
the general taxpayer. 

The Consolidated Kail Corporation, the chosen instrument for em- 
bodying the reform visualized in point (2) of this statement, has not 
yet accumulated enough experience to indicate what final sums may 
be involved in connection with points (5) and (6). Moreover, there is 
still the highly-vexed (and much-litigated) question of what treat- 
ment is to be accorded security-holders in predecessor railroads. But 
the principle is there, and it is essentially new : in at least a large por- 
tion of the Uited States, a prime instrument for public policy toward 
unprofitable branch lines must be Federal financial aid. Kegulation 
can no longer carry the load, unassisted, which was once thrust upon 
it in a different economic environment, because many of the carriers 
themselves can no longer carry the load. 

The present position may be summarized as follows: 

(1) The Interstate Commerce Commission has full legal powers to- 
prevent branch-line abandonment. It obtained these powers in 1920,. 
when it was assumed that such abandonments would be confined to 
timber railroads in cut-over areas and other clear examples of eco- 
nomic obsolescence. 

(2) Today's abandonment problem is a systems problem, in three 
senses : it involves systems, such as the predecessors of ConKail, in 
which the parent body cannot provide the strength to support losing 
branch lines or any other parasites; it involves the basic question, 
which cannot be answered by any one railroad, of how the operation 
of each branch line affects both the net benefits obtained and the 
profitability of the entire railroad system ; and it involves the issue of 
how to confront abandonment proposals which are so widespread 
that they amount to quasi-systems rather than isolated ganglia. Neither 
the organization of the industry nor the experience of the Interstate 
Commerce Commission is appropriate as a basis for a rational answer 
to any of these three questions. 

(3) The regulatory process probably gets in the way of true collec- 
tive bargaining. Most branch lines are of limited, if positive, national 
interest. In many cases, the closest approach to an economic solution of 
the branch-line problem would lie in negotiations between the carrier, 
on the one hand, and both the shippers and the communities located 
along the line, on the other. Kegulation, as such, can provide no funds 
for continued operations. Kegulation, as such, cannot force communi- 
ties to subsidize such service. The Interstate Commerce Commission 
may have a role in such negotiations, as essentially a referee. This role 


is, in fact, explicitly recognized in Section 802 of the 4-K Act of 1976,. 
dealing with arrangements for continuation of service on low-density 
branch lines. But the financial aspects of such arrangements are ta 
be administered bv the Secretary of Transportation (Section 803; 
see also Sections 804, 805, and 50V515). Under present-day circum- 
stances in the railroad industry, the main effect of regulation on the 
bargaining positions of those who supply rail services vis-a-yis those 
who demand or desire the services is to weaken the bargaining posi- 
tion of the former and to add to the difficulty of working out qukl-jno- 
quos in the form of sliipper or community guarantees of minimum re- 
venue in return for railroad guarantees of minimum service. 

{c) Car service 

The primary difficulty with respect to the timely and adequate pro- 
vision of rolling stock' is not the fault of the Interstate Commerce 
Commission; in fact, the Commission's unenviable role has been to 
make the best of an economically unsound situation. 

The unsoundness of the car service situation in the American rail- 
road industry is a product of the Balkanization of the industry. The 
situation is complicated by an imbalance of traffic movements which is 
largely the result of the commercial geography of the Unitod States. 

Balkanization has had both positive and nejiative ill effects. 

On the negative side. Balkanization has obscured the important rela- 
tionship in freight car supply, which is the relationship between the 
railroad system and individual shippers. The important rental price 
for a freight car is the demurrage charge, that is, the rental paid by 
a shipper while a car is on his siding or at his disposal. The important 
relationship of this price is to the frequency and reliability of rail- 
road switching service. Yet the Balknnization of the industry has 
caused most attention to be paid to internal charges, in the form of 
per diem rates to non-owninir railroads. 

On the positive side. Balkanization has led to an artificial separa- 
tion of the railroad rate from both the railroad per diem and the 
railroad demurraire charge, "\An-iat mattei^ first to a railroad is the 
relationship of all receipts, however derived, to all costs. T^Hiat mat- 
ters next to a railroad is the relatio7iship of the structure of receipts 
to the stmcture of costs. In particular, freight cars should be ex- 
pected to earn their keep over time, but without being a source of 
extraordinary profits. As long as some railroads originate dispropor- 
tionately large amoimts of interlined traffic and others terminate 
large amounts while originating disproportionately small amounts, the 
economic signals of cost and return which ou£rht to govern the size 
of the freight car fleet are blurred. A railroad which specializes in traf- 
fic solely over its own line is in a position to make rational decisions 
with respect to ownership of freight cars. Xeither a railroad which 
originates but does not terminate a large volume of traffic, nor a rail- 
road which is in the opposite position, can make rational decisions on 
its own. Eational decisions would renuire both the correct car-rental or 
per diem rate and the correct rate for the movement of traffic. Such 
decisions are a function of the entire tariff and not of any one seornent. 
Shippers are concerned with the total bill and not exclusively with any 
one separate component. 

In the absence of an institutional stmcture which would generate 
rational economic decisions by itself, the Interstate Commerce Com- 


mission has had to struggle with problems which could only be com- 
pletely solved if the Commission had complete managerial control 
over the railroad industry and the money available to convert this 
managerial control into active investment policy. The Interstate Com- 
merce Commission has had the legal power both to establish car 
service rules and to issue car service orders since 1917.^^ It has exer- 
cised these powers, often in conjunction with the Association of 
American Railroads or its predecessor organizations, on a continuing 
basis with respect to car service rules and sporadically, in response to 
what have appeared to be acute freight car shortages, with respect to 
oar service orders. In both cases, regulatory (or trade association) 
intervention has in effect amounted to the partial substitution of 
rationing for the allocational effects of a price system. 

The combined effects of these interventions, and of the rigid pricing 
rules and fragmented patterns of freight-car ownership which made 
them necessary, have been summarized in the following criticisms : 

... The consequences of this failure to enlarge the role of the per diem system 
and utilize the forces of the market to establish car-rental rates have, however, 
been most perverse: (1) rate inflexibility has prevented appropriate recognition 
of the locational and seasonal variations in demand; (2) preoccupation with 
historical costs of car ownership has tended to obscure the relationships between 
prospective daily ownership costs of newly-acquired freight cars and the per 
diera rate as a crucial element in the investment-decision process; (3) the con- 
flict between the goals of returning cars to their owners and of achieving opti- 
mum utilization of a national freight car fleet has not been resolved ; (4) a vast 
bureaucracy in the ICC and the AAR has been created to develop and administer 
ever more elaborate and complex rules and formulas and exceptions thereto ; 
and (5) there has been a steady substitution of compulsion for agreement, i.e., 
rules and car-hire charges which were once established by the voluntary action 
€f the railroads have been subject in recent years to imposition by the ICC." 

This is a classic summary of the self-reinforcing character of any 
rationing system : over the long run, it tends to exacerbate the very 
conditions which were the cause of its creation. The passage also 
clearly brings out the special possibilities for sub-optimization which 
such a system can create in the Balkanized environment of American 
railroading. Given the fragmentation of freight car ownership, the 
need for through movements of freight cars over many different lines, 
and the imbalance among American railroads both in finances and in 
the relative importance of originated traffic, imposition of rigid pric- 
ing systems is bound to produce both a need for supplementary reg- 
ulations and something like the contours that present regulations have 


Regulation in the field of transportation is in no sense as all-embrac- 
ing, either with respect to coverage or substantive content, as regula- 
tion in the field of public utilities. Even where regulation is universally 
acknowledged to be important, as is true for intramodal and inter- 
modal rate relationships, a case can be made (and has been made, by 
the Interstate Commerce Commission), for the proposition that reg- 
ulatory intervention at most affects the tail of a large animal. In other 
areas where regulation might be expected to be important, such as that 

«' John Richard Felton, "Freight Car Shortages in Grain Transport : An Institutionalized 
Disequilibrium" (mimeographed), Lincoln, Nebr., May 1974, p. 89. 

*« Felton, op. cit, p. 91. This passage may perhaps be overly-enthusiastic in Its refer- 
■ence to "a vast bureaucracy." 


of controllincr the earnings permitted to regulated firms, full applica- 
tion of public utilit}^ standards has never been achieved or even really 
attempted. For railroads as an industry, such application might even 
be economically impossible. 

The conclusions to be drawn from this partial and uneven state of 
transport regulation might include the following : 

(1) The squeaky wheel gets the grease. In the case of railroad 
regulation, this means that the industry's attempts to retrench m 
response to a shrinking market (e.g., by branch line abandonment) 
are the occasion for long-drawn-out and laborious proceedings before 
the Interstate Commerce Commission. In the case of motor-carrier 
regulation, this means that, in a growing industry where entrance is 
more attractive than exit, most of the time of the Interstate Commerce 
Commission is taken up with cases involving applications for operating 

(2) Pricing decisions are unrelated to investment decisions. Pricing 
decisions by regulated carriers on regulated commodities are legally 
subject to complete regulatory surveillance. Yet the implications of 
these decisions for rates of return and investment policies are rarely, 
if ever, considered in rate cases involving specific commodities, and 
only partially considered — for railroads — in general rate increase 

(3) It is doubtful whether a regulatory body can ever be a planning 
body. Among otlier reasons, planning requires access to funds in order 
to make plans effective; it requires an affirmative and forward-looking 
view of market processes and developments: it demands a capacity for 
ulthnate executive action which a ''quasi-judicial'' regulatory body 
does not have. But, in transport regulation, it is impossible to develop 
even a consistent approach to planning, or a harmonious environment 
within which planning can take place. Too many of the relevant vari- 
ables are beyond the control, or even the purview, of the regulators. 

(4) Keguiation is no substitute for money. If the regulated industry 
is capable of earning true monopoly profits, then a regulatory body 
can — doubtless over the protests of economists — divert some of these 
profits for the benefit of what it believes to be particularly meritorious 
or impecunious classes of consumer. If the surplus profits cannot be 
generated, or are dissipated through competition among quasi-regu- 
lated firms, then the redistributional function of regulation must 
wither away. Xo one can redistribute what never existed in the first 

In sum, the salient feature of the Interstate Commerce Commission 
is not how much it does, or how much it does well or badly, but how 
much it does not do, does not want to do, and in the last analysis cannot 
do for either political or economic reasons. For example, no one has 
ever proposed complete regulation of all private trucking. Economi- 
cally and administratively speaking, an attempt at such regulation 
would make no sense regardless of legal requirements or regulatory 
practices. In the absence of regulatory control over private trucking, 
regulation of every conipetitive mode of transport must under any 
circumstances be selective and uncertain in its total economic effects. 

^»Thls situation may chanee somewhat as the I.C.C. decides on cases In conformltv with 
the new standards of the 4-R Act of 1976. 

Impact of Current Eegulation: Costs and Benefits 

Most economic discussions of the importance and effects of regula- 
tion come from economists avIio tend to maximize the cost of regu- 
lation. In attempting to quantify the effects of regulation, they 
naturally tend to come up with numbers to match. The present discus- 
sion is more skeptical. If this skeptical position is to be supported 
further, it must first proceed through a critique of the very high 
numerical values for the cost of regulation which are now current in 
the economics profession. 


These estimates have become almost a minor industry for trans- 
portation economists. The easiest way out would appear to be to begin 
with a source which combines within itself a number of advantages: 
it is compendious; it is latitudinarian, in the sense that it presents 
widely-separated minimum and maximum estimates ; and it has prob- 
ably been more frequently quoted than any other estimate or combina- 
tion of estimates. The relevant information from this source, a paper 
by Thomas Gale Moore, is contained in Table 3. 

[In millions of dollars] 

Estimate of loss 
Type of loss 

Inefficient use of mode: 

Common carrier truck 

Private trucks 

Rails .•_. 

Water carriers ._ _ _.. 

Pipelines _ 


Traffic shifted to alternate mode: 

Trucks to rails 

Water carriers to rails 

Pipelines to other modes 


Traffic not carried.. 

Total estimated loss 3,775 5,560 8,890 

1 Not estimated. 

Source: Thomas Gale i^/loore, "Deregulating Surface Freight Transportation," ch. 3 of Almarin Phillips (editor), Promot- 
ing Competition in Regulated Markets, The Brookings Institution, Washington, D.C., 1975, table 3-2, p. 71. 

The sources cited in the Moore table are the following: The low estimate is based on Merton J. Peck, "Competitive Policy 
for Transportation?" in Almarin Phillips (editor), Perspectives on Antitrust Policy, Princeton University Press, 1965, pp. 
261-255. Medium and high estimates are derived from Robert W. Harbeson, "Toward Better Resource Allocation in Trans- 
port," Journal of Law and Economics, 12 (October 1969), 322-334. 

As Moore points out, "all the estimates given here are very crude." ^° 
So there would be no point in quibbling about even a range of specific 
numbers. But there are certain matters of principle lying behind this 
kind of estimate, and this kind of estimation, which must be brought 




1, 400 

1, 700 








3, 400 




'. 200 








00 Thomas Gale Moore, "Deresulating Surface Frei?:ht Transportation," Chapter 3 of 
Almarin Phillips (ed.). Promoting Competition in Regulated Markets, The Brookings 
Institution, Washington, D.C., 1975, p. 72. 

to the forefront if any progress is to be made in establishing the 
dimensions of the regulatory problem : 

(1) It has been a truism of railroad economics since the 1840s that 
the railroad industry is a decreasing-cost industr}^ in the sense that 
a heavier volume of traffic per mile of road will typically cut unit costs 
both in the short run (more effective use of existing way and struc- 
ture, possibly more effective use of train crews, etc.), and in the long 
run (in normal circumstances, adjustment of way and structure, via 
double or multiple-track operation or otherwise, does not require an 
increase in cost equivalent to the increase in volume which can be 
accommodated). It has also long been a truism of economics that 
resources are put to their best possible use, when viewed from the 
standpoint of a given industry if not necessarily from the standpoint 
of the economy as a whole, if marginal price equals marginal cost: 
i.e.. if the economic value of the extra resources which must be em- 
ployed to provide an additional unit of a good or service are worth, 
in other uses, just the price they receive in this particular use. Other- 
wise, potential buyers ready and willing to pay more than the extra 
costs of providing the extra service are cheated of an opportunity to 
enjoy a net benefit over and above the cost of serving them. There- 
fore, the argument concludes, optimum allocation of resources re- 
quires the avoidance of ''dead weight loss'': output should not be so 
reduced by high prices that consumption witli a greater unit value 
than the offsettincr unit cost is forced out of the market. 

The trouble with the marginal-cost argument in any decreasing 
cost situation is that, if every price is equal to marginal cost, total 
revenues will necessarily fall short of total cost. Hence the dilemma: 
if an enterprise in this situation is forced to sell every unit of output 
at exactlv what it costs to produce the last (or next) unit, it cannot 
avoid deficits and presumably ultimate oronomic extinction. The solu- 
tion to the problem whirh was first worked out in the 1050? may be 
summarily described as follows : if pricino- at marginal cost will result 
in a loss, adopt the "inverse elasticitv rule'': charge customers with the 
least elastic demands the highest prices relative to marginal costs, and 

The first defert of the "economic loss'' cnVnlations rited in the !Moore 
conti'ibution. then, is that thev conr^entrate on marcfinal conditions and 
iimoT'e total conditions. If all railroad rates were at marginal cost, 
who would pick up the check for tlie remainder? 

(2) An answer to this arGrument can be approached by the asser- 
tion that present-day American railroads should be put through the 
di'Jinvestment wrinirer. "Wliat better wav than thronorh encouraging 
or pven forcing them to char.cre rates ennal to maririnal cost? 

The trouble with this art<=wer is that it is deficient both as a matter 
of theorv and as a matter of practice : 

(ti) As a matter of tlieory. the "decreasing costs with ofreater 
densitv*' characteristic of railroads hns nothino- to do with 
growth, stairnation, or decline. Much railroad capital is special- 
ized capital, as to function as well as geocrraphical location. 
Therefore, as economic tides recede, railroad investment without 
significant alternative uses will be left hiorh and drv. But this is 

"^ For example, see Maroel Bniteux. In "Sur la grestinn des monopoles publics astreints 
a r^quilibre budg^taire," Econometrica, 24 (January 1956), 22-40, especially pp. 38-39. 


a separable economic issue, having nothing to do with the "den- 
sity" economics of the railroad industry even when it is holding 
its own or expanding. 

(b) In the railroad industry, there is no practical way to 
respond to shrinking demand by infinitesimal reductions in track 
gauge, or weight of rail, or crew size, or the size or weight of 
rolling stock. ^Yhen costs decrease with increasing density, they 
are typically lumpy during volume decline as well as growth. 

(3) When attention is shifted from the marginal cost argument 
for railroad charging, with all of the analysis confined strictly to the 
railroad industry, to a marginal cost argument on an intermodal basis,, 
the new problem is that railroads and motor carriers cannot possibly 
perform the same function. Indeed, motor carriers have more service 
flexibility than railroads, just as railroads have more cost flexibility 
than motor carriers. Motor carriers not only have more service flexi- 
bility; in general, they also have a service advantage. But this ad- 
A^antage varies route by route, commodity by commodity, and even 
shipment by shipment. Under some circumstances, motor carriers may 
even have a disadvantage. Any estimate of "loss through regulation" 
which is based on comparative rail and truck costs infringes a funda- 
mental economic rule : don't directly compare apples and oranges. 

(4) As has been stressed in previous portions of this discussion, 
motor carriers regulated by the Interstate Commerce Commission 
are in some respects less competitive with rail than they are with 
other motor carriers, or than those other motor carriers are with rail. 
Tlierefore the existence of keen intramodal competition does not 
necessarily imply a major regulatory influence on intermodal 

(5) Railroading and motor carriage are both indutries whose value 
added represents an unusually high percentage of total revenues. Sinre 
both are sellinjr services which require the direct use of labor, both 
are relatively labor-intensive and, conversely, little dependent on the 
outside purchase of goods and services. In labor-intensive industries, 
both the productivity and the rates of pay of members of the labor 
force are major ingredients in determining the j)rices the industry 
will charge for its products. 

To take the railroad case first: it would be most difficult for an 
economic historian to conclude that railroad wage rates, railroad 
fringe benefits, or railroad operating rules could be traced, even in- 
directly, to the existence of the Interstate Commerce Commission or 
the thrust of any of its decisions. A more plausible argument would 
be that both railroad regulation and railroad unionization have been 
more or less simultaneous products of a common set of economic and 
social factors. But to attempt to translate this vague generalization 
into any kind of correlation between regulation and unionization 
would require an exceptionally vivid imagination. The main, if not 
sole, influence of the I.C.C. on railroad labor has occurred, rather 
adventitiously, in connection with I.C.C. jurisdiction over rail merger 

The trucking case is not so open-and-shut. Indeed, a significant 
study of the relationship among unionization, regulation, and wages 
in the motor carrier industry reaches the following conclusions : 

. . . ICC regulation of entry and price has increased the employers' ability 
to pay. . . . Hence, the excess iirofits of the cartel have been expropriated by 


the IBT [International Brotherhood of Teamsters] and divided among its 
members. . . .^^ 

The author of this passai^e points out that ''wages and major fringes" 
grew steadly from 5o.2 percent of motor carrier annual revenues in 
1945 to 65.0 percent in 1965 (old definition), and from 53.4 percent 
in 1965 to 59.8 percent in 1970 (new definition) . Over this same period, 
1945-1970, '-the trucking wage" grew at an annual rate of 5.4 percent, 
as against a growth for yearly income in all private industry of 4.5 
percent. In 1945, "Annual income in motor freight and warehousing'' 
v\-as 12.9 percent higher than annual income in all private industry; 
in 1970, it was 39.8 percent higher.^^ 

Therefore the hypothesis that I.C.C. regulation of motor carrier 
rates has in Qf^LCct "become a conveyor belt for delivery of benefits to 
members of tlie Teamsters' Union cannot be dismissed out of hand. 
This assertion could be supported by perusal of any motor carrier 
general rate increase case. Wage increases are treated as independent 
data, with I.C.C. response in terms of rate adjustments expected to 
be the dependent variable. 

But even this interesting hypothesis may prove too much, for the 
following reasons : 

(a) As far as the intercity trucking industry is concerned, the great- 
est strength of the Teamsters' Union lies among over-the-road drivers. 
Pay of the over-the-road driver is an especially important component 
of total cost for long-distance truckload shipments, which are precisely 
those shipments which ofi'er the most direct competition with rail. 
Yet this is also the stronghold of the owner-operator. And it is hard 
even to visualize what rigid wage-and-hour stipulations for owner- 
operators miglit mean. Conversely, the largest regulated motor carriers 
tend to have high percentages of less-tlian-truckload business, and 
therefore high percentages of tei-minal. billing, and other such costs 
which may be affected little if at all by the activities of the Teamsters' 

(b) Xot all regulated motor carriers are unionized. Conversely, 
very many members of the Teamsters' Union are employed in con- 
nection with private or otherwise non-recrulated motor carriage. There- 
fore it is impossible to establish a one-to-one correspondence between 
regulation, on the one hand, and the Teamsters' I'nion. on the other. 

(c) As for the data concerning relative incomes: the motor carrier 
iiulustrv has been one of the healthiest American industries, and 
among the faster-growing, since 1945. The financial health of the inter- 
city poition of the industry may have been improved more by the 
Interstate Highway Program than by all the activities of tlie Inter- 
state Commerce Commission put together. And productivity .<rains 
have been especially marked for over-the-road drivers — where Team- 
ster membership is concentrated — as compared with terminal, pickup- 
and-delivery, and office workers. 

These comments serve to qualify acceptance of the proposition that 
the Interstate Conmierce Commission has become the inadveHent 
captive of the Teamsters' Union, and to cast doubt on the proposition 

"2 James E. Annable. .Tr.. "The ICC. the IBT. and the Cartolizatlon of the Araerlcnn 
1 ruokinjr Industry." The Quarterly Review of Economics and Business, l.*? (Summer, 1973), 

»■> Ibid.. Table o. p. 40, and Table 6, p. 43. 


that reofiil cation, by itself, has significantly raised motor carrier costs 
by permitting higher wage rates. 

(6) The problem of economic methodology connected with these 
estimates may be summed up by considering the Moore heading, 
"Static Welfare Loss." ^* 

The technique of comparative statistics is a perfectly justifiable 
method for establishing first approximations to the economic effects 
of policy measures in a dynamic economy. But these first approxi- 
mations can nevoi' be accepted as final conclusions. For example, con- 
sider the following passage : 

. . . Assume . . . that rail rates would fall suflBciently to maintain rail traffic 
in the absence of regulation — that is, rail prices would fall 20 percent as truck 
rates fell 20 percent. The net impact of these reductions in rates is to increase- 
truck traffic by 18 percent. Since total class I and class II truck revenues were 
$9.6 billion in 1908, the total change in traffic revenue would be approximately 
$1.7 billion, and the total welfare loss would be $175 million. This estimate, how- 
ever, is a conservative one since it does not allow for any increase in shipments 
by rail.^ 

But since the operating ratios of I.C.C.-regulated motor carriers 
(which make no allowance for return on capital investment) are 
typically in a range of 94-95,^^ it is not obvious how this 20 per cent 
reduction might be achieved except through massive reorganization 
of the motor carrier industry. On further investigation, it appears 
that the 20 per cent estimate is based on actual experience in the mid- 
fifties with motor carrier rates on various agricultural commodities 
(fresh and frozen poultry and frozen fruits and vegetables).®^ 

The American Trucking Association has not done a particularly 
convincing job of assuming its burden of proof on the question : why 
should the experience with deregulation of these agricultural com- 
modity rates twenty years ago not be generalized to the entire indus- 
try ? ^"^ But to leap from experience with limited groups of agricul- 
tural conmiodities to all regulated motor carriers is, at best, to leap in 
the twilight. What is required, and has never been really attempted, is 
a careful inductive study of just how representative (or unrepresent- 
ative) the deregulatory experience in agriculture has been. 

(7) If the rates reduction cannot come out of profits, it must oper- 
ate through pressure toward lower costs. One way this might occur 
is by reducing excess capacity. Measures of excess capacity in regu- 
lated transport, or in transport in general, have taken two forms: 
absolute,^® and relative.^°° 

84 Op. cit., p. 70. 

8= Op. cit., pp. 70-71. 

8«E.g., the Interstate Commerce Commission 88th Annual Report to Congress, 1974, 
Table 10, p. 125, shows operating ratios for Class I inter-city motor carriers of property 
as follows : 1971, 94.05 ; 1972, 94.41 ; 1973, 94.75 ; for the depressed years 1974 and 1975\ 
the I.C.C. 90th Annual Report, 1976, shows rates of 95.25% and 95.01% (Appendix E, 
Table 10, p. 146). 

»7 Moore, op. cit, p. 60 ; Ann F. Friedlaender, "The Dilemma of Freight Transport 
Regulation," The Brookings Institution, Washington, D.C., 1969, p. 74. 

83 See, for example, Allan C. Flott, "The Case Against the Case Against Regulation," 
I.C.C. Practitioners' Journal, 40 (March-April 1973), 281-290. The main thrust of the 
Flott rebuttal relates to an alleged unsatisfactory character of unregulated truck service 
for seasonal livestock movements. Yet regulation has never distinguished itself — nor, until 
1976, had to concern itself — with the problem of seasonal pricing in specialized transport 
movements. Even now, the "seasonal flexibility" emphasis is still restricted to rail move- 
ments, which have largely disapi .ared with respect to livestock. 

88 See Friedlaender, op. cit., especially Table 4.4, p. 85, and Table 4.5, p. 88. 

i«> Moore, op. cit, Table 3-1, p. 63. 


(a) Absolute measures of excess transport capacity, whether they 
attempt to attribute the excess to reoulation or any other influence, 
suffer from the following inherent defects : 

(i) For railroads, there has always been a substantial margin 
of excess capacity on branch lines and on less-favored main lines. 
The decline in passenger transportation has tended greatly to 
increase the excess in many parts of the system. This excess can 
only be reduced by abandonment and cannot easily be absorbed by 
routing shipments over the "excess" facilities, since a shipper is 
interested in purchasing delivery at a destination and not in ton- 
miles or car-miles per se. Therefore, if one proceeds beyond the 
more obvious branch-line problems, the question of how regulation 
contributes to this excess must be handled with extreme care. 

Moreover, if "excess'' is measured in terms of rolling stock, the 
fact must be faced that two of the most dynamic markets for rail- 
road services in the last decade or so have involved special rack 
cars for new automobiles, and special unit trains shuttling between 
coal mines and electric powerplants. In both cases, and in many 
others like them, empty car mileage is bound to hover around 50 
percent of the total. 

Finally, the geography of the United States, together with tlie 
special comparative advantages of rail transport, combine to pro- 
duce unbalanced rail movements. Eail traffic to and from southern 
New England and the major northeastern metropolitan areas is 
very unbalanced, for reasons having almost nothing to do with the 
Interstate Commerce Commission. 

(ii) For trucking concerns, the inherent geographical imbal- 
ance of rail may also exist — but not necessarily with respect to the 
same regions. Regulated regular-route motor carriers concentrate 
on shipments of manufactures; irregular route bulk carriers must 
often employ very specialized equipment. Both may produce 
"built-in" geographical ini})alance in freight movements. 

For reasons of the loading characteristics of different commodi- 
ties, the same type of comment holds for data on "Actual reve- 
nue ton-miles as percentage of potential revenue ton-miles : aver- 
age load per truck." ^°^ The common-carrier trucking industry 
tends to attract light-loading commodities, in part because it is 
not suited to bulk commodities such as coal and many building 
materials. Light-loading commodities "cube out" before they 
"weigh out." 
(h) Relative measures of excess capacity — involving comparisons 
between the position of regulated and unregulated carriers — are obvi- 
ously impossible for the railroad industry because the industry is en- 
tirely regulated. But they can be employed for the motor carrier 

Moore's conclusion from the data available to him is implicitly 
contained in a footnote to his Table 3-1 : "Private capacity shortfall'' — 
"The percent of private capacity that could be saved if utilization level 
were the same as for regulated vehicles." ^^^ The footnote tells nothing 
alx)ut the "percent of private capacity" which could, in fact, be saved 
if regulation were eliminated. AVhen one consults the source of the 

101 Fripdlaender, op. cit., Tables 4.5, p. 88, heading of column 2. 
1^2 Moore, op. cit., p. 63. 


Moore table — an article by E-dward Miller ^°^ — one discovers that the 
emphasis is not on the comparison of regulated and private trucking 
contained in the Moore table, but on the absolute aspects of regulated 
carriage itself : 

. , . Thirty-eight percent of for-hire multiple unit vans return empty [in the 
for-hire trucking industry] even though such equipment is unspecialized. This 
seomi=: to far exceed the amount of empty mileage that would result solely from 
directional imbalance in the traffic. . . . The large amount of empty mileage ob- 
served supports, although it does not prove, the allegations that regulation has 
led to inefficiency in trucking/"* 

This statement points the wav toward the necessity to collect and 
compile more useful official statistics on all motor carrier transporta- 
tion.^°*^ But it is clearly not meant as conclusive in itself. 

(8) Transportation has a special appeal for econometricians, and 
for quantitative economists in .Q-eneral, because this sector of the econ- 
omv vields such a wealth of statistics. 


(i) Many of the current statistics are of a vintao^e, in terms 
of the kind of nuestion they can be used to answer, which may be 
indicated bv the prevalence among them of the description! 
"Steam railroads." 

{ii) Eailroad statistics are, in terms of coverage, complete; 
m.otor carrier statistics refer — at most — to carriers regulated by 
the Interstate Commerce Commission (not CA^en most of the 

{Hi) The Census of Transportation, which is the most reliable 
current source of data with respect to many aspects of total truck 
movements, and of the competitive relationship between rail and 
motor carrier, is not truly a "census"; it leaves out practically 
all motor carriage except that of manufactures, and is based on 
a statistical sample even of this sub-category. 

(w) Transportation involves specific geographic movement, 
and specific geographic movement involves specific origins and 
specific destinations. The Civil Aeronautics Board generates 
masses of data with respect to the origins and destinations of 
movements by commercial airline ; the Interstate Commerce Com- 
mission generates no specific data with respect to origins and 
destinations of freight movements by rail. When the I.C.C. pub- 
lished the 1 percent waybill sample it at least provided state-to- 
state information on rail freight movement. Once the waybill 
sample was shi fted to the Department of Transportation, even this 
information ceased to be publicly available. 

Therefore, a major criticism of all of the attempts made so far 
to place a monetary value on the costs of regulation is that they 
are based on underlying statistics which simply will not support 
the weight of the argument (whether pro or con). 

(9) As a guide to policy determination, the process of quantita- 
tively estimating "losses from regulation" suffers from both a theo- 
retical and a pragmatic defect. 

103 "Effects of Regulation on Truck Utilization," Transportation Journal, 13 (fall, 1973), 

10* Miller, op. cit, pp. 13-14. 

^""' A first step in this direction has I een taken by the Interstate Commerce Commission 
itself in Bureau of Economics and Bureau of Openitions, "Emptv/Loaded Trucic Miles on 
Interstate Highways During 1976," April 1977. 


(a) Theoretically, the "comparative statics'' analysis used in most 
analyses of losses from regulation is impossible to fit into any time 
period. All modes of transportation, like other portions of the econ- 
omy, are involved both actively and passively with a dynamic econ- 
omy. A major change such as deregulation would clearly alter the 
dynamics of regulated transport firms — but this alteration in dy- 
namics, however beneficial it might be to all concerned, cannot be 
forecast by economic exercises which start bv assuming one allocation 
of transportation resources and end by assuming a very different allo- 
cation. Meanwhile, the most interesting economic questions — and the 
interesting policy questions — are concerned not with the characteristics 
of two states of rest, but with the direction, speed, and rate of accelera- 
tion or deceleration of economic movement. 

(b) Practically, the "comparative statics*' approach neglects the 
applicability to the railroad industry of Lord Keynes" aphorism, "In 
the long run, we are all dead." There is a serious question as to the 
applicability of any portion of standard long-run economic analysis 
to the railroad industry as a whole. There is, of course, an accompany- 
ing serious question as to the long-run vialulity of large segments of 
the industry without direct public assistance. That is true now — with 
regulation. Until and unless these questions have somehow been re- 
solved, long-run conclusions based on assumptions of movement from 
one equilibrium position to another quite different one assume, possi- 
bly without justification, tliat there is anotlier quite different equilib- 
rium to move to. 

In addition, for botli rail and motor cai'riage. tho all-or-none ap- 
proach implied by an attempted comparison of present regulation with 
complete deregulation is both politically unrealistic and economically 
beyond the reach of customaiy economic analysis. Mainstream eco- 
nomics is marginalist economics. It is at its best in examining the direct 
and indirect effects of relatively minor changes in the status quo. Radi- 
cal economics, whether of the left or the right, is not really concerned 
with such humdrum issues as deregulation of surface freight transport. 
In both cases, to begin with the assumption of large changes in a micro- 
economic area is to do an injustice to either traditions and analytical 
technique, on the one hand, or ideological fervor and Messianic vision, 
on the other. 


Capture theories 

One large family of approaches can be by-passed at the outset. This 
family represents a highly-miscellaneous group of spokesmen who are 
convinced that the Interstate Commerce Conmiission (like all other 
commissions which have been in inisiness for any length of time) is 
simply a captive of the industries it supposedly regulates. It has been 
pointed out that the 'capture- theory is 

. . . Espoused by an odd mixture of welfare state liberals, muckrakers, 
Marxists, and free-market economists. . . }°^ 

Whatever the detailed authenticity of the ^capture' theory, it is wide 
of the mark as an explanation for the attitude of the Interstate Com- 

if« Richard A. Posner, "Theories of Economic Regulation," Bell Journal, 5 (Autumn, 
1974), 335. 

83-944—78 5 


merce Commission toward discontinuance of passenger trains or 
abandonment of branch lines ; it does not provide a very convincing ex- 
planation for I.C.C. policy toward general rail requests for rate in- 
creases; and it may even be self -contradictory if it is used as an 
explanation of I.C.C. policy toward more than one transport mode on 
occasions when interests of different modes are in conflict (as they 
w^ere, for example, in the Ingot Molds case) . 

In its extreme form, the 'capture' theory can produce odd results. 
For example, Chapter 6 of ^''Ttie Interstate Conwnerce Omission^'' is 
devoted to "Highway Safety." ^^^ Early in the chapter, the comment 
is made, 

Though the direct enforcement of federal trucking-safety regulations was 
transferred to the Department of Transportation, the Interstate Commerce Com- 
mission still retains a potentially powerful indirect sanction against unsafe prac- 
tices. As a condition of acquiring and keeping an ICC "operating certificate," a 
carrier must maintain its "fitness," of which safety is supposed to be one 
criterion ^^ 

Subsequently, the horrendous weekly schedule of a steel hauler is 
quoted, verbatim, from the hauler's own diary."^ Immediately after 
these quotations from the diary, we are told : 

Men forced to haul the often overweight payloads of bulk steel often operate 
on schedules like the one above. Their equipment is frequently substandard be- 
cause as owner-operators they cannot always afford to make repairs right 
away. . . }^ 

Therefore the Interstate Commerce Commission stands indicted be- 
cause of a purported indirect control (via certificates of convenience 
and necessity) over a group of carriers (owner-operators) who are 
beyond its control entirely. 

Thus the "capture" theory merges with the idea that "any stick is 
good enough to beat the Interstate Commerce Commission with." 
Neither of these approaches points the way toward responsible policy- 

^Vhat, then, can be said in connection with qualitative estimates of 
the costs of regulation ? 

The first point worthy of note is that any realistic examination of 
regulatory costs should be on a net basis. Presumably regulation con- 
fers benefits. On whom ? 

Who henefits from regulation? 

Apparently not, in general, the shippers. A relatively recent exten- 
sive questionnaire on the subject was sent to 920 potential respond- 
ents — one half to transportation executives of corporations, one half to 
corporate presidents — in July, 1973. The response rate was on the low 
side, as it generally is with voluntary questionnaires : 32 per cent for 
transportation officials, 13 per cent for general executives, and 22 per 
cent over-all. A majority of both groups favored less regulation of 
both rail and highway carriers. The majority was larger for non- 
transportation than for transportation executives.^^® 

io<' Robert Fellmeth, "The Interstate Commerce Omission" (Introduction by Ralph Nader), 
Grossman Publishers, New York, 1970, pp. 190-222. 

i"7 Ibid., p. 192. 

108 Ibid., pp. 201-203. 

ifoibid.. p. 208. 

"" Donnld V. Harper and Tames C. .Johnson, "The Shipper Views Deregulation of Transpor- 
tation." I.C.C. Practitioners' Journal, 41 (March-April 1974), 320, 321, 329. Since ques- 
tionnaires were ori.jrinally dispatched on the basis of a random sample of corporate executives 
taken from Mood.v's, small businesses must have been under-represented, in terms of both 
number and total economic importance, and the largest corporations also under-represented, 
in terms of economic importaoce. 


Not the Association of American Railroads, at least with respect to 
the regulatory situation as it stood before the passage of the 1076 
amendments to the Interstate Commerce Act: included in the AAR 
list of "Burdens placed on railroads by the government" was "Cost or 
[sic] rate regulation ($500 million a year) ."' "^ 

Apparently benefits are conferred on both regulated trucking com- 
panies and members of the Teamsters' Union. At least, both groups 
have been adamant in their opposition to regulatory change with re- 
spect to Part II of the Interstate Commei'ce Act (that dealing with 
motor carriers), and have so far manifestly had an influence in pre- 
venting such change. 

The Interstate Commerce Commission has tended, until very re- 
cently, to regard ^'regulatory reform" as a device for introducing more 
and stricter regulation. 

Where, then, stands that amorphous mass — the general public? 
The first thing to note is thnt. with respect to regulation of freight 
transport, it is hard for tie public to stand anywhere. The only aspect 
of freight transport regulation which directly affects them is that 
relating to household goods' moving and storage; and it is probably 
no coincidence that this is an area which has given rise to extensive 
and continuing complaints to the Interstate Commerce Commission. 
Otherwise, shippers and consignees must stand in as proxies for the 
public, because freight transport only indirectly yields final consumer 
How well do shippers and consignees represent the geneial public? 
At long as the present regulatory stnicture continues, pi'obably as 
well as can be expected. Tracing the ultimate incidence of changed 
freight rates on commodities such as steel, coal, woodpulp, or sand and 
gravel is surely beyond the competence of the economics profession in 
the foreseeable future. Even if it could be determined that individual 
members of t]ie public were being nickeled-and-dimed to death by ex- 
cessiv^e freight rates or needlessly inferior service, it Avould be excep- 
tionally difficult to collect any portion of these nickels and dimes in 
order to fight a case through the Interstate Commerce Commission 
and the courts. 

But is this a reassuring conclusion, or simply a devastating com- 
mentary on the present regulatory structure ? 

Probably some of both. In the* days of railroad monopoly, efforts to 
moderate raih'oad rate discrimination could be justified as a means 
of protecting the public against lonor-run exaction by either or both of 
fi) monopolistic railroads who might emerge from the final Armaged- 
don of predatory competition: (ii) monopolistic shi}>pers who might 
drive actual competitors out of business, and intimidate potential com- 
petitors, by preferential rail rates. As the economic significance of rail- 
roading has waned, the interest of shippers in "orderly'' rate relation- 
ships is less likely to reflect any ultimate interest of consumei-s. At the 
rate structure level, the public jias far less to gain from regulation than 
it once rlid, and as much to lose in the form of regulatory complacency, 
inflexibility, and attempts to preserve obsolete economic relationships. 
From the standpoint of rate level, in the case of railroading, ship- 
pers may provide better representation for the public than the public 

-f,>i^;^^<'f'"i*'"t o^ Stpphon Alles. Presirlent, Association of Anipricnn Rnilronds "Rnilroad^— 
!.><.). hpann^s before the Snhrommittee on Surface Transportation of the Committee on 
pp'ToToOD ■ ^'''^^^^' ^^^^ Congress, 1st Session, part 4. G.P.O.. Washington, D.C., 1975, 


could provide for itself. Shippers presumably have some grasp of the 
economic situation of the American railroad industry, and of the 
longer-run significance of the present state of railroad"^ finances. The 
American consumer movement has not had mucli direct experience with 
an industry whose main problem is not merely "Should consumers 
pay?'' but rather "Is there any way they can be cajoled into using the 
service ? " 

For the motor carrier industry, the public interest is probably more 
important on the supply side than on that of demand. Here is an in- 
-dustr^/ which offered tremendous scope for new entrepreneurial talent 
^fty years ago. As it has matured, this scope would in any case prob- 
ably have decreased to some extent. But the whole process of apply- 
ing for certificates of convenience and necessity is designed to daunt 
new entrepreneurs who have the ability and willingness to graduate 
from the ranks of owner-operators or enter from other fields. 

T)oes mention hring neto bene fits from, regulation of freight wove- 

This brings us to the question of how freight transport regulation 
relates to such macro-economic issues as inflation and unemployment. 

As long as all the labor aspects of freight transport are kept in 
a compartment which is practically beyond the reach of traditional 
recfulation, there can be no immediate and direct connection. 

With respect to indirect effects, it is likely that regulation impedes 
price increases by regulated carriers in the first stages of accelerating, 
demand-pull inflation. In particular, it tends to impede rate increases 
for products w^hich are subject to sudden upsurges in demand — such 
as that for bituminous coal in consequence of what happened to oil 
prices beginning in the autumn of 1973. But, in this latter case, it is 
not clear that the consumer even obtains any short-run protection. The 
result may sim.ply be to increase the economic rents obtained by the 
fortunate possessors of the scarcer commodity. 

Conversely, under the peculiar conditions of cost-push inflation and 
general economic slackness which prevailed in the United vStates in 
1970 and for much of 1974-75, regulation probably tends to hold prices 
up even in the short run. 

In the lonsfer run, regulation must answer the question: liow are 
the regulated carriers to achieve greater cost-effectiveness? Here the 
problem is not just the fact that regulatory bodies have no control 
over waire rates or fringe benefits, nor is it even the fact thnt regula- 
tion necessarily creates pockets of economic inefficiency if the alter- 
native would be a genuinely competitive industry. The essential prob- 
lem is that there is no way for any regulatory body actively to promote 
efficiency. Under Section 15a of the Interstate Commerce Act, the 
Commission is empowered to judge whether there is "honest, economi- 
cal, and efficient management." But it has no powers to create or even 
encourasre such management. And, from recent imofficial diatribes of 
Commission Chairmen on the subject of northeastern and midwestern 
railroad managements, one would be forced to the conclusion that the 
Commission believes that questions of economic efficiency are com- 
pletely beyond its^ control. Therefore, in the longer run,^ regulation 
can confer no public benefits in the way of lower factor costs or greater 
factor productivity. The most it can do is to prevent such possible 


gains from bein^ translated into excess profits (for the railroad in-- 
dustry. read: "niore nearly normal profits''). And tlie difnculty witlr- 
this is that it either tends to inhibit carriers from striving for such- 
profits, or forces them to watch them being frittered away in compul- 
sor}' losses. 

As for employment in regulated industries, there is really no way 
of determining what the general effect of regulation might be. On the 
one Jiand. both minimunrrates and economically-needless restrictions 
on service (of the kind commonly found in the regulation of motor 
common carriers) would tend to increase employment over the level 
tliat would prevail for that traffic volume in the absence of regula- 
tion. On the other hand, higher traffic volumes should be achieved, 
for regulated and unregulated carriers combined, in the absence of 

Other possible heneiits from regulation 

These appear to be minimal indeed. 

Perhaps the quickest way to establish this point is to review a talk 
given in 1974 by then Commissioner (now Chairman) A. Daniel 
O'Xeal. Commissioner O'Neal spoke from the double vantage-point of 
the Interstate Commerce Commission and previous experience at the 
highest staff level with the Senate Commerce Committee. He was 
speaking to a forum organized by a group with an outstanding impu- 
tation for advocacy of free and unregulated markets. Therefore one 
would expect the case for the benefits of regulation to be put with 
particular vigor and cogency. The net product was along the follovring 

rrobably the most important function of regulation is to eliminate unjust 

Comment: Historically, very tnie; semantically. always true, be- 
cause who is opposed to the elimination of "unjust" practices of any 
kind? But. for the motor carrier industry, the possibility for any sig- 
nificant discrimination would be practically nil for tru<"kload move- 
ments because private trucking stands ready to take over freight sub- 
iect to high rates. Moreover, this possibility is of a second order of 
magnitude even for less-than-truckload shipments. The argument 
r'ould more aj^propriately be turned around : for motor carriage, regu- 
lation has probably created more discrimination — just or unjust — 
tlian it has eliminated. For railroads, the original discrimination issue 
i=i still very much alive. But the question again arises: what is the 
economic justification for preventing firms which can scarcely stay 
alive from charginc: what the traffic will bear? And if there is such a 
justifictaion. what is the further justification for imposing the burden 
of a range of other policy decisions on the railroads? 

. . . Some argue . . . that actually the marketplace is a much better regulator 
of quality than is a federal bureaucracy. It seems to me this theory is signif- 
icantly weakened if one analyzes industries that are unregulated in the sense we 
mean it and where the quality is open to criticism. The automobile industry and 
the meat-processing industry are obvious examples here."^ 

1^2 "Xo riamor for Doregrulation : Should Therp Bp — ". in James C Mlllpr ITT. Persnpc- 
tivps on Fpdpral Trnnsnortafion Policv. Aniprican Enterrri<;p Institutp for Pn'ilic Policv 
Rp-parr-h. Washington. D.C.. 1975, p. 191. 

"3 Ibid., p. 102. 


Cprrmient. — To continue with the automobile analogy, regulation of 
minimum rates tends to recreate the position summed up in Henry 
Ford's famous comment, "They can have any color they want as long 
as it's black." In the absence of regulation, Ford lost a large part 
of his market to other firms that offered more style than the Model T. 
But a standard rate begets a standard quality, even for buyers who 
prefer Lincolns or bicycles to Model Ts. 

. . Transportation is important not only as a means of moving people and 
goods from point to point, but as an instrument of social and economic develop- 
ment. For example, in a pure market situation elimination of branch lines may 
make good sense, but where it is thought useful to maintain the vitality of agri- 
cultural areas or smaller cities, elimination of branch lines may be intoler- 
able. . . .^* 

Comment. — This statement essentially refers to the period before the 
development of modern highways. Moreover, if branch lines are to be 
maintained, the only sure way to maintain them is the same as the 
only sure way to have them built originally — by public contribution. 

. . . Deregulation may create dense populations in city centers and social 
and economic dislocation. . . ."^ 

Comment. — Seventy-five years ago, quite possibly; today, it would 
be next to impossible to predict that deregulation would have any ap- 
preciable effect on the relative distribution of economic activities as 
between large cities and small. Given available unemployment data, 
sustenance for 'dense populations in city centers' might be a worthy 
objective. But there is little that the presence or absence of an Inter- 
state Commerce Commission can do, any longer, to affect this situation. 

. . . when the railroads attempted to make major rate readjustments on 
agricultural commodities with ICO acquiescence in the 1920s, the country ex- 
perienced a classic case of congressional intervention. . . . Congressional action 
took the form of the Hoch-Smith resolution. . . .^^^ 

Com.m,ent. — Note the date ; note the economic position of agriculture 
then and now ; note the economic position of railroads, then and now. 
Note, finally, that Hoch-Smith was not a law but a "resolution." 

... This . . . suggests another social reason for moving traffic at less than 
resource cost — the reason being the need to redistribute income. The lower prices 
that occur on food help low-income or moderate-income groups. Higher rates on 
high-value commodities may take something away from the rich and from in- 
dustry, but they help keep food moving at lower prices.^" 

Comm^ent. — A mass of impossible objectives for any agency regu- 
lating freight transport, tied together by threads of ancient history. 
There is no connection between rates on any type or geographical 
movement of freight and the general income levels of shippers or final 
consumers. The country already possesses an elaborate food-stamp pro- 
gram — if, indeed, subsidized food consumption is, per se, a desirable 
objective. Low transportation rates on unprocessed foodstuffs cannot 
be enforced in the trucking industry, because these are exempt from 
regulation anyway. If they are enforced in the railroad industry, the 
result may be mainly to raise land values at remote locations. And 
"high- value commodities," whatever they may be, cannot possibly bear 
any specifically- discernible relationship to the rich or to anyone else. 

"* Ibid., p. 192. 
lis Ibid., p. 104. 
"« Ibid., p. 195. 
"» Ibid., p. 196. 


There would be no point in extending these comments beyond the 
length of the ori<zinal O'Xeal presentation. But of all the benefits 
presented by a distinguished spokesman to what could only have 
been a skeptical audience, there is not one which has any unambiguous 
and distinctive economic content of current significance. Therefore 
tlie skeptical conclusion must be : if this is a typical apologia for In- 
terstate Commerce Commission activities, then the clock has indeed 
stopped somewhere between 1887 and 1920. 

In short, the case against present regulation as exemplified by the 
activities of the Interstate Commerce Commission cannot be estab- 
lished by the presence of quantitatively-identifiable costs. The com- 
plaint relates to the absence of qualitatively-identifiable benefits. The 
Interstate Commerce Commission is not corrupt, or captive, or ma- 
licious, or someliow designed to play havoc with optimum allocation 
of resources. Instead, it has lost the sense of the present-da}^ economy, 
and of how present-day freight transportation — regulated and unreg- 
ulated — fits into that economy. 

The Current Economic Circumstances Surrounding Eegulation 

degree of modal transport monopoly 

There is nothing which even approaches a present equivalent to the 
position of American railroads before 1920. The railroads face sharp 
competition from trucks, pipelines or water carriers in virtually every 
category of freight."^* 


As measured by total number of firms, railroads are considerably 
less competitive than they have been in the past. As measured by num- 
ber of firms serving specific routes, they have on the average become 
considerably less competitive in the northeast — due not only to the 
creation of Penn Central and the Chessie System, but also the distres- 
sing series of events which led to the formation of ConRail. If regu- 
lation were to be totally eliminated, tliere is no reason to believe that 
railroads would behave as pure competitors — if, indeed, anyone can 
imagine how a "purely competitive*' railroad might be expected to 
behave. Nor would unregulated railroads have tlie empire-building 
incentives to extreme competition which often prevailed both before 
and after the advent of interstate regulation. 

One familiar point must be raised before the railroads are dis- 
missed as a series of "live and let live" oligopolies. This pertains to 
their interrelationships : "Some 50 percent of all shipments (accounting 
for about 70 percent of rail ton-miles) are interlined. . . ." ^^^ If all 
present regulatory legislation were wiped off the books, this presum- 
ably Avould carry away the Reed-Bulwinkle Act. Interlining requires 
noorotiations as to both through rates and rate divisions. The result- 
ant joint rates must often be in harmony with single-line rates. Thus, 

■'^'^ Note that this statement is not meant to relate to the "Market Dominance" pro- 
visions of Section 202 of the 4-R Act of 1976 — as discussed, for example, by the I.C.C. 
in : 'The Impact of tlie 4-R Act Railroad Ratemakinpr Provisions." op. cit., pp. 22-4.3. 

"^Alexander L. Morton, "Is There an Alternative to Regulation for the Railroads?" in 
James C. Miller III (ed. ), op. cit.. p. 29. "Interlining" refers to the transfer of freight cars 
from one railroad to another at junction points. 


in rate as in other matters, there is a necessary degree of complemen- 
tarity which would color whatever competition might develop. In 
view of this complementarity, the loss of competition which might 
otherwise be anticipated from certain types of mergers could, in fact, 
be minimal. Meanwhile, effective policy with respect to rate bureaus 
can hardly be visualized apart from policy with respect to mergers. 

In the motor carrier industry, even the very large number of regu- 
lated carriers may not adequately indicate the present degree of active 
competition for truckload traffic on major routes. Exempt and private 
carriers, plus owner-operators who may hire themselves out to com- 
mon carriers as well as others, give the industry a great potential for 
competitive flexibility even in the presence of regulation. It is question- 
able what control of certificates of convenience and necessity accom- 
plishes on major routes, aside from making a limited contribution 
toward ease of enforcement of minimum prices. And Interstate Com- 
merce Commission sanctions for the rates filed by motor carrier rate 
bureaus must tend toward reducing the spectrum of competitive results 
(in terms of differences in services rendered, as well as differences in 

But if parts of the motor carrier industry are at least competitive as 
measured by the number of regulated carriers and other parts are 
affected by the spillover of non-regulated carriers and non-regulated 
carriage, neither standard can be applied to the markets in which the 
largest firms operate. These firms have obtained systems advantages 
through the possession of ver}^ extensive operating rights, and would 
be somewhat protected from extensive competition even in the absence 
of continuing regulatory constraints on potential competitors. But 
there may already be large firms with enough mileage of routes so 
that opening up all markets to all firms would even increase the com- 
petition among these relative giants. This question needs separate 


This was a great fear of the early railroad days. It is still present 
today. But, for railroads, the prize typically Would not be great 
enough : driving an important railroad which was previously solvent 
out of business by predatory rate cuts would impair the solvency of 
practically any would-be predator; and driving a railroad already 
near insolvency out of business would involve unnecessary trouble in 
obtaining what could be had for the asking. 

As for the permanent elimination of motor carrier competition by 
either a motor carrier or a railroad : given the extreme mobility of 
the industry's major capital investment, this would appear to be a.<? 
difficult a feat as there is in American economic society. Resrulation 
which pins motor carriers to specific routes, origins and destinations, 
gateways or commodities does not protect against predatory competi- 
tion ; it invites it. 


The great age when railroads filled in the maps of the Middle West 
and West is now a century behind us. At most, they can still have 
considerable influence on the location of specialized activities such as 


coal mines. Motor carriers tend toward complete neutrality in their 
effect on location: what matters is the infrastructure (roads). These 
are a product of government planning and expenditure, and not of 
the actions of common carriers, or of regulatory bodies. Tlie only form 
of freight transportation which probably retains some power to in- 
fluence specific industrial location is inland water transportation. 
Here, again, the influence is a result of public investment policy. 



From the price and service standpoint, it is difficult to cast motor 
carriage in this role. Railroads still have some possibilities both for 
aggressive discrimination on behalf of certain shippers and for yield- 
ing, passively, to certain shipper demands. But a glance at any map 
which shows plant locations for industries which specialize in the 
fabrication of bulk commodities indicates that these industries are 
particularly drawn to waterways, although they may also be drawn 
to locations which enjoy some degree of rail competition. The con- 
trolling factor with respect to most plant locations which are attracted 
to railroads is not likely to be either present rates or present service, 
but the trend of service and the degree of probability that the railroad 
will still be there a decade or so hence, A branch-line which is the 
potential object o,f an abandonment proceeding is an exceptionally 
poor candidate for attracting railroad-oriented industries. 


This idea has already cropped up at intervals throughout this 
di^-T'ussion. It is time at last to describe what it means. 

Subsidization occurs whenever anything is sold to any buyer in 
exchange for a total revenue which is less than the total (or avoidable) 
cost of supplying that particular buyer. The loss incurred on this 
buyer ma}' be recouped from a number of places — or it may not be 
recouped at all. If total revenues are adequate, then this recoupment 
involves "cross-subsidization." If they are not. then so-called "cross- 
sub.-idization" may actually be only" a partial offset to "seller sub- 
sidization" (i.e., subsidization by the seller). 

Regulated motor carriers are perfectly capable of earning normal 
returns, or — in the presence of regulated minimum prices and entry 
controls — substantial abnormal returns. Therefore forced cross-sub- 
sidization as a result of regulatory decisions is entirely possible in 
theory. But it is unlikely to be important in practice. Motor carriers 
not only avoid most of the costly features of the railroad branch-line 
pi'oblem; on truckload movements, they have no distinctive terminal 
problem, and on less-than-truckload movements local terminal ex- 
penses can be cranked into over-all rates much more easily than 
Avith rail. Moreover, motor freight carriers share with rail freight 
carriers a great deal o,f service flexibility, and obviously have no direct 
incentive to give good service or even to continue any service to small, 
remote.^ customers in out-of-the-way places, even in the presence of 

Therefore the whole idea of cross-subsidization at regulatory behest 
tends to fall between two stools in freight transportation. Railroads 


could cross-subsidize if they were prosperous enough ; in fact, however, 
services which do not yield avoidable costs are often if not generally 
subsidized by the railroads' owners (or creditors) rather than by other 
traffic. Regulated motor carriers are prosperous enough; but one 
reason why they are this prosperous is the extreme difficulty of forcing 
them, by regulatory measures, to cross-subsidize. 

Possible Alternatives to the Current Regulatory Scheme 

Here there are two very different situations. 

the railroads 

The positions of American railroads is in an unprecedented state 
of flux both with respect to regulation and with respect to government 

The unfeasibility of any major cross-subsidization of railroad traf- 
fic under modern conditions was first confronted by Congress when 
railroads were given the option of shifting their passenger service to 
AMTRAK. The necessity for subsidy if intercity rail passenger traf- 
fic is to be maintained has now found its way into the annual budgetary 

A further need for public assistance, as well as a kind of reorganiza- 
tion which went far beyond anything the Interstate Commerce Com- 
mission had ever done or was legally empowered to do, lay behind the 
passage of the Regional Rail Reorganization Act of 1973 which pro- 
vided for direct government subsidy to finance the maintenance of 
branch-line service. Most recently, the Railroad Revitalization and 
Regulatory Reform Act of 1976 has introduced the possibility of major 
changes both in regulatory practice with respect to railroads and with 
respect to the financing of the industry. 

Even when they are considered without reference to the financial 
provisions of the legislation of the last decade, the proposals for reg- 
ulatory reform in the Act of 1976 are the most important to be con- 
tained in any railroad legislation since the Act of 1920. Moreover, they 
point — in the long run — away from the 1920 Act. The Act of 1^20 
tightened regulation of railroads in a number of important ways. The 
1976 Act loosens it by permitting: rate variation without I.C.C. ap- 
proval within a "zone of reasonableness." The Act of 1976 also breaks 
entirely new ground in different directions : e.sf., Section 206 is headed 
"Rate Incentives for Capital Investment" (both the idea of a rate 
incentive and the idea that rates mav have some connection with in- 
vestment are new, and thorou^rhly commendable) ; Section 205 is ap- 
propriately headed, "Adequate Revenue Levels"; and, in particular. 
Section 202 adds the followinof as part of a new sub-paragraph (b) 
to Section 1(5) of the Interstate Commerce Act : 

. . . No rate which contributes or which would contribute to the going concern 
value of such a carrier shall be found to be unjust or unreasonable, or not shown 
to be just and reasonable, on the ground that such rate is below a just or reason- 
able minimum for the service rendered or to be rendered. A rate which equals 
or exceeds the variable costs (as determined through formulas prescribed by the 
Commission) of providing a service shall be presumed, unless such presumption 
is rebutted by clear and convincing evidence, to contribute to the going concern 
value of the carrier or carriers proposing such rate. ... In determining variable 
costs, the Commission shall, at the request of the carrier proposing the rate. 


determine only those costs of the carrier proposinc: the rate and only those costs 
of the specific service in question ... no rate shall be found to be unjust or unrea- 
sonable, or not shown to be just and reasonable, on the ground that such rate ex- 
ceeds a just or reasonable maximum for the service rendered or to be rendered, 
unless the Commission has first found that the proponent carrier has market 
dominance over such service. . . . Nothing in this paragraph shall prohibit a rate 
increase from a level which reduces the going concern value of the proponent car- 
rier to a level which contributes to such going concern value and is otherwise 
just and reasonable. . . . 

An effort to expound the exact mennino: of this passage so soon after 
its appearance in the statute books would almost be classified as 
contempt of court. 

It would appear, however, that Conofress intends to reverse Ingot 
Molds — on the assumption that "contribute to going concern value" 
and "from a level which reduces the going concern value ... to a level 
which contributes to such . . . value" are to be construed as referring to 
rates which are already above marginal cost, or at least rising toward 
marginal cost, and on tlie further assumption that "determine only 
those costs of the carrier proposing the rate" means exactly what it 
says. For, on these assumptions, the new Act appears to sweep away at 
least some of the confusions between marginal and aA'erage cost, and 
the tendency toward so-called "umbrella rate-making," which wore 
evident in Ingot Molds. It would seem that the a])propriate cost for the 
relevant carrier is now the law of the land, as a check on any vagrant 
regulatory or judicial tendency to wander toward the inappropriate 
cost of the irrelevant carrier. 

It would also appear that the Interstate Commerce Commission 
must now seek out residual areas in which railroads still have indoj'jen- 
dent power over markets and hence over rates (the "market domin- 
ance" provision), and can no longer take its previous lurisdiction over 
rates for irranted. In view of a headline in the Traffic World — 
"Market Dominance Issue Taken to Appeals Court" — it would be pre- 
mature to ofo beyond this hesitant statement."^ 

In addition to the resrulatory reforms introduced by this passage, the 
list could be continued almost indefinitely: through provision for sea- 
sonal rates, provision for relating demurra<^e charges to costs of enuip- 
ment, thi'ouofh the institution of "l)racket" rate-making fin the fii-st 
year, plns-or-minus seven per cent), and so on. For present purposes, it 
is less important to repeat legislative provisions than it is to stress cer- 
tain fundamentals which seem to be emerging with respect to the fu- 
ture of railroad regulation : 

(1) Regulation is no longer to be considered in isolation from 
the financial condition of American railroads, or from their access to 
whatever capital is believed to be required if they are to perform the 
transportation services expected of them. 

(2) Tlie fallacy of "cross-subsidization" in railroading is being 
faced: if losing services are required of the railroads, the tendency 
is to reimburse at least part of the loss directly [AMTRAK, branch- 
line subsidies]. 

1^9 Traffic World, Xov. 29. 1976. cover summary of contents. The t^\t which amplifio^ thl.s 
headline has this to say: "Nation's railroads ask U.S. Court of Apppnls for District of 
Columbia to enjoin and set aside ICC definition of 'market dominance' to extent found 
unlawful, and accuses ICC of restlne -ts definition on premises that are not rational or 
pvidencp-piipportPd." For an exr.lanation of the present position of the I.C.C, consult 
"The Impact of the 4-R Act Railroad Ratemaking Provisions." op cit. 


(3) The two-headed kind of management situation which strong 
and inflexible regulation creates — with one head really capable only 
of saying "Xo*' — is to be reduced. Words like "incentive'' appear; 
ideas like "going-concern value'' are incorporated in legislation ; doors 
are opened which will permit railroads, if they wish, to engage in more 
flexible and market-oriented comm.ercial policies. 

(4) Transcending these and other specific points, the Congress has 
obviously become convinced that the railroad industry as a series of 
privately-owned corporations is in a genuine crisis. The Congress 
appears to be further convinced that both the nature and the propor- 
tions of this crisis are such as (a) to make regulation according to 
guidelines which were essentially laid down in 1920 completely an- 
achronistic in today's setting: (b) to make regulation, liowever mod- 
ernized, quite inadeouate to solve problems involving rail transporta- 
tion which will yield only to money; (c) to make regulation concede 
at least some ground, in some areas, to direct decision-making by 
railroad managements. 

(5) All of this constitutes not just modernized regulation, but a 
imodernized approach to the present-day railroad industry. Of course. 
Itliis modernized regulation and this modernized approach are not only 
a long way from 'deregulation'; in the short and even medium run, 
they will probably move in an opposite direction. Almost every sen- 
tence in the Act of 1976 will require full and formal consideration by 
she Interstate Commerce Commission; almost any action by the Com- 
mission as a result of such consideration is likely to be appealed to the 

^Courts ; many of the issues raised by the Act are so fundamental that 
only the Supreme Court can pass on their legal validity and precise 
Jeo-al meaning. So, for a believer in deregulation, the Act of 1976 can 
at best be described as a tactical retreat : each hoped-for step toward 

..fl<^^'^^'^dation is likely to be preceded by two steps back. 

Therefore, it is still entirely possible that the economic dynamics of 

fjiB railroad industry will not wait for the procedures established in 
the new Act. But, for the moment, an economist must remain in a state 
of anticipation pending the new decisions which the Interstate Com- 
merce Commission is compelled to make in the process of converting 
the generalizations of the Act of 1976 into specific regulatory 
change. ^^^^ 


For regulated motor carriers, the present outlook is: more of the 
same (i.e.. stalemate). It would be easy to become cynical about this 
situation, and to allege that both trucking concerns and their em- 
ployees have acquired so much political power that the law will con- 
tinue to be iust what they want it to be. Since this approach is so 
plausible, a balanced outlook requires due scrupulousness in present- 
ing the other side : ... 

(1) An important motivating factor in the radical changes wliK-h 
are now taking place, and being prepared for, in railroad regulation 
was obviouslv the financial collapse of practically every railroad in the 
Northeast which did not have large and reasonably stable revenues 
from bituminous coal traffic. There is no such crisis in the motor 

119a :MeanwhilP. railroads seem to have been very cantious about takincr advantaere of 
the 4-R Act flaring the first IS months or so of its validity. (See I.C.C., "The Impact of 
the 4-R Act Railroad Ratemaking Provisions," op. cit. 


carrier industr}-. On the contrary, in most y^ars this industry earns 
rates of return ^Yhich are among the best for any industry in the 
United States. So the internal pressure for refonn is simply not there 
in trucking. 

(2) Although this writer has never seen a 'second-best' case pre- 
sented for the regulated motor carrier industry, such a case can be 
made.^^° Critics of motor-carrier regulation argue that, in its absence^ 
the motor carrier industry would be highly if not purely competitive. 
That, proponents of continued regulation might rejoin, is precisely 
the point. Every Tom, Dick, and Harry with the price of the down- 
payment on the cheapest of second-hand trucks would be able to enter 
the industr}^ and in the process he and his like would drive down 
trucking rates so that the country would start to consume too much 
in the way of trucking service. Also, in the process, a low-income en- 
clave would be established due to excessive ease of entry. Consumers 
would not only consume too much in the way of trucking service, in 
some absti-act sense, but specifically, they would consume too much in 
the way of trucking service relative to rail service. For both financial 
and economic reasons, lail roads cannot reduce rates to the level of their 
own marginal costs. Pui'ely competitive truckers, however, could be 
expected to do precisely tliis. Hence undue diversion of traffic from rail 
to motor carrier. 

This is not a theoretically contemptible argument. The trouble is 
that it probably exairgerates the impact of present regulation of motor 
carriers on raili'ojnl traflic As has been stressed ]n-eviously. there are 
ample grounds for skepticism about the economic pi'essurcs exerted 
1\V the competition between I'egulated common carrier trucks and rail- 
I'oads, u.ndei- pi'esent-day conditions (not those of I0o5. or 1055). But 
the main point is that some of the fledgling carriers of 1035 are now 
iho economic giants of 1078. In many areas where entry miirlit be easy, 
in the total abseu'^e of regulation, it already is easy, in the presence 
of regulation, for private and exempt carriers, and leased 
owner-oj)erators. Conversely, the regulated motor carrier industry, in 
that portion for whicli i-egulation is most extensive (i.e., regular-route 
common cari'iers of general commodities) already has a limited num- 
ber of very large firms whose economic strength is mainly a product 
both of pi-esent economics and of past regulation. If the man on the 
street were free to step into a truck and haul anything he wanted any- 
where he wanted for any client he could find, the efTect on United 
Parcel Services or Consolidated Freight ways or Roadway or Over- 
nite or Yellow Freiirht or any of the most important carriei'S who are 
relevant to this port of the discussion would be minimal. The largest 
carriei's tend to have the most far-flung operating rights — and among 
the hi fiber percentairos of less-tlian-truckload traffic. The effect on 
American railroads, however, would perhaps be perceptible. (In view 
of the present broad sweep of unregulated trucking in truckload 
lots, even this new competitive situation mijzht yield no results beyond 
what has already happened. The main impact on railroads would 
pi'obablv arise fi-om the new freedom of private carriers to accept 

'-'* For the orlprinni definition of the term 'second-best' ns It Is used In economics, anrl 
« theoretical description of how it mav be applied, see R. G. Llpser and Kelvin Lan- 
caster, "The General Theory of Second Best." Review of Econcvmic Studies. 2P. (1955-56^), 


(3) If there is no crisis from the standpoint of owners and workers 
in the regulated motor carrier industry, there is, it must be admitted, 
no perceptible crisis from the standpoint of shippers. Part of the 
reason may be that the regulated, regular-route, motor carrier spe- 
cializes in less-than-truckload shipments — and "small shipment" is by 
no means synonymous with "small shipper." Thus, from Eichard 
Stuart, of the Traffic Division of American Home Products Corpora- 
tion, we find that 

. , . large volume small shippers are able to exercise leverage to gain re- 
duced rates. . . . 

the solution is simple — a combination of individual shipments and local 

All a shipper must do ... is combine enough small 2,000-pound shipments to 
obtain truckload rates and ship them en masse to central distribution 
points. . . .'=" 

In short, a large shipper may avoid regulatory problems with small 
shipment rates simply by minimizing small shipments, and relying 
on motor carriage where motor carrier rates and service are relatively 
most affected by competition and relatively least affected by regula- 
tion — i.e.. for movements in truckloads. 

This still leaves us with that legendary supposed beneficiary of mo- 
tor carrier regulation — the small shipper of small shipments from 
small towns. He is in no position to testify before the Interstate Com- 
mission in his own behalf, and not in much of a position to become 
a member of a small-shipments trade association which will do the 
job for him. Yet the differential disadvantages which motor-carrier 
regulation imposes on him are supposed, somehow, to be of maximum 
assistance to him. 


(1) Although quantitative measures of the cost of present regula- 
tion tend to exaggerate its importance, this provides a basis for criti- 
cizing the quantitative measures, not for supporting or endorsing pres- 
ent regulation. 

(2) More fundamentally, traditional economic analysis is simply 
not well-equipped to measure the true costs of regulation, which con- 
sist of introduction of needless delays, of needless inflexibilities, and 
of needless substitutions of legal fictions for economic realities. 

(3) As compared with these very substantial true costs, the benefits 
of regulation must be consigned very largely to the past. Under pres- 
ent-day conditions, regulation of intercity freight movements either 
undertakes tasks which regulation cannot successfully perform, or 
undertakes tasks for which there is no economic justification. 


(4) The role of the railroad in the American economy has changed 
drastically since regulation was introduced. When the Act to Regu- 
late Commerce was passed in 1887, some railroads were actually very 
prosperous and most railroads appeared to be potentially so. The 
industry was growing rapidly; its economic advantages relative to 
other forms of transportation were either already unassailable or in- 
creasing; and railroad influence on the course of general economic 

^ Traffic World, June 24, 1974, p. 81. 


development and on the state of competition throuorliout the economy 
was of first importance. The public attitude toward the industry, and 
toward its place in the economy, was rightly molded by these under- 
lyinof considerations. Today the railroad industry is scarcely arrow- 
ing, if growing at all : its economic advantages relative to other forms 
of transportation are either under assault or decreasing: and the rail- 
road influence on the course of general economic development and 
on the state of com])etition in other industries is generally not great. 

(5) The significance of this historic transformation for regula,- 
torv policy is that the public attitude which first found expression 
on Federal statute books in 1887, and was finally and fully spelled out 
in 1020, has been a relic of the past and not a ornidepost toward the 
future at least since 1945. As A]\rTRAIv and ConEail both serve 
to indicate, the most pressing problems of today's railroad industry 
are not problems toward the solution of which any re.frulatory body, 
operating under anv conceivable legislation, can possibly contribute 
niiythinof substantial. In extreme cases, the only choices are provision 
of public funds (with or without public ownoi-ship) or abandonment. 
In intermediate cases, attempting to force railroads to provide internal 
subsidies for particular commodities, or communities, or regions 
amounts in effect to forcing railroad security-holders to provide this 

(6) Fortunately, raili'oad regulation has alreadv been moved a long 
wav from its dead-center of the period from 1020 right up to the 
1070s. The Railroad Revitalization and Regulatory Reform Act of 
1076 is the most important railroad regulatory legislation since the 
Transportation Act of 1020. Its general long-run thrust is toward de- 
regulation: but. in granting additional powers of judgment and 
interpretation to the Interstate Commerce Commission, its immediate 
effect is likely to be to increase the economic significance of regula- 
tion. However, the real importance of the Act lies in the fact that it was 
addressed to the railroad problems of 1976, not the railroad problems 
of 1020 or 1887. 

(7) The 1076 Act performs the further useful function of putting 
everyone on trial: the railroads, the Interstate Commerce Commis- 
sion, and the Congress. The "Regulatoiy Reform" aspect of the Act 
ci-eates a new environment within which the American railroad indus- 
try is given opportunities it did not have before to become more effi- 
cient and more commercially-oriented. It gives the Interstate Com- 
merce Commission a new opportunity not only to reappraise existing 
policies but also to demonstrate that this reappraisal can be accom- 
plished in a reasonably short time. The Interstate Commerce Commis- 
sion of a previous era essentially disregarded its public function by 
failing to take the Act of 1920 seriously except to the extent that the 
Act enhanced its own regulatory powers. The Act of 1076 gives it 
another and pcrliaps final chance. The Act also gives Congress a 
unique opportunity to monitor the practical impacts of new provisions 
wliich appear, on paper, to permit genuine and irreversible changes 
in the character of the American railroad industry. 

(8) The pragmatic issue presented by the Act of 1976 is therefore : 
sliould the Congress wait until some returns are in before pressing for- 
ward with further new rail legislation? The answer would appear to 
be '*Yes,'' unless an irrevocable decision is made to dispense with regu- 


lation of rail freight entirely. In view of both the commodities and 
areas in which railroads still have a dominant market position, and of 
the structural changes which have just occurred (ConRail) or appear 
to be possible in the fairly near future (reports of various discussions 
relating to merger) , this would not seem to be an appropriate time for 
eliminating regulation. A year from now, the impacts of the regulatory 
reforms of 1976 and of various possible merger proposals should begin 
to emerge. These impacts should provide the basis for subsequent legis- 

(9) It would be premature to suggest what such legislation might 
contain. But, as is already evident from the financial commitments 
which the Federal government has made to the railroad industry 
during the 1970's, the tone of this legislation will have to be affirmative 
in the sense of reflecting more than a Federal policy dedicated solely 
to "permitting free operation of market forces.'- ;^'^arket forces have 
never operated with textbook freedom in the railroad industry. In 
some countries, they were never even permitted to operate from the 
outset. In the United States, national control over such operation was 
decided on as early as 1887. We are now in a period during which the 
future of private ownership of railroads is highly precarious in im- 
portant regions of the country and is nowhere assured. Under such cii*- 
cumstances, the essential regulatory commandment — "Thou shalt 
not" — is no longer appropriate; but complete reliance on "market 
forces," without public guidance or assistance, would be equally 

Congress has already faced up to the fact that changes in regulatory 
policy are but one phase of changes in public policy toward railroads. 
(For example, note the title of the Railroad Revitalization and Eegu- 
latory Reform Act.) What the future seems to hold in store is a further 
blending of new and affirmative public policy, public money, and re- 
form of public regulation. 

(10) There is one area in which further regulatory reform could be 
hiitiated without a prior "wait-and-see" attitude toward the Act of 
1976. This is the area of deregulation of railroad rates for commodi- 
ties whose shipments are not subject to rate regulation when they take 
place by other transport modes! Thus, agricultural commodities are 
exempt from regulation if they move by motor caiTier or baro-e; their 
rates are regulated if they move by rail. To allay fears, regulation of 
maximum rail rates could be retained. But, below these maximum 
rates, railroads should be given complete freedom to compete with 
other modes, and incidentally to offer special bargains on the transpor- 
tation of agricultural commodities. 


(11) When one turns from rail to motor carrier, the obsolescence of 
present regulation is equally obvious. But the reasons are very differ- 
ent. There was a strong economic as well as political case for regulat- 
ing railroads in 1887, and in 1920. The remains of this case still retain 
some validity, both with respect to commodities which are more or less 
captive to rail transportation (e.g., most long-distance overland move- 
ments of coal), and with respect to control of possible rate discrimina- 
tions betAveen individuals and between places as well as between 


There never was any such case for the regulation of motor carrier. 
The motor carrier industry operates relatively small service units over 
publicly owned facilities. Its route structure and geography could be 
almost completely flexible in the absence of regulation. Moreover, 
there is only one portion of one relatively minor branch of the motor 
carrier industry which provides a possible argument for direct con- 
sumer protection. This branch is that of moving and storage. Finally, 
there is no really effective way for the Interstate Commerce Commis- 
sion to exercise"^ detailed and continuous control over motor-carrier 
service standards. Most freight need not, and should not, move on 
vehicles which are held to rigid passenger-type schedules; in the 
absence of such schedules, it is hard to attach a specific meaning ta 
the idea of "service." Therefore the idea that regulation can protect 
small shippers and small towns by forcing cross-subsidization is an 
idea with little possible content. Even if such a policy were desirable, 
no regulatory body could effectivelj^ carry it out. 

(12) At the interface or frontier between rail and motor carriage, 
I.C.C. regulation can accomplish relatively little, by way of the estab- 
lishment of minimum rates or the exercise of any other legal powers, 
to allocate shipments between rail and truck. Hence the near-disap- 
pearance of what was probably the main reason for the passage of the 
Motor Carrier Act of 1935. 

('13) If vnUvoad reasons for motor carrier regulation have prac- 
tically vanished, motor carrier reasons have been created, in large part 
by the regulation itself. 

First, there is the demonstrable fact that, on the average, motor- 
carrier rates of return have exceeded rates of return in the average 
unregulated industry and even further exceeded rates of return in 
other i-egulated industries. 

Second, there is the demonstrable fact that the regulated motor car- 
rier industry is, in fr^^npi'^b unionized. The most important union in 
the industry, the Teamsters, has managed to enforce a Avage scale 
which has steadily risen from not much above the national average, 
in 1945, to very considerably above the national average at present. 

Third, there are the A'erv special interests of reofulated common car- 
riers of general commodities who have obtained extensive operating 
riirhts by purchnse or by direct grant from the Interstate Commerce 
Commission. Within the regulatory framework, (1) some of the largest 
carriers have obtained the difTerential advantage (from the service 
and demand side; not necessarily just from the cost side "I of very ex- 
tensive operating rights whose value would be reduced if competitors 
were allowed to enter all markets freely; (2) this differential advan- 
tage has little or nothing to do with competition involving anv other 
transport mode. To take the extreme case of the Inrirest I.C.C. -voo-ii- 
lated common carrier: United Parcel Services cannot r)os?iblv be 
deemed a competitor of any water carrier. Its relationships to rail- 
roads are essentially complementarv, due to U.P.S. use of railroads 
for manv shipments, and hardly at all competitive. 

Fourth, there is the fact that shippers of commodities by re.o^ulated 
motor carriage haAT had over forty years to become accustomed to the 
regulatory system. Additionally, the largest users of motor carriage 
are also typically in the best position to resort to private carriage if 
they believe this to be advantageous, or to bargain for commodity rates, 

83-944—78 6 


or to insist on the best service for their shipments. They are also in 
the best position to influence regulatory policy. 


Given the obsolete character of present motor carrier regulation — 
indeed, probably of all motor carrier regulation — and given the ob- 
durate resistance of many interests connected with motor carrier regu- 
lation to any substantial change in the present situation, is there any 
way out of the apparent impasse ? 

Step One has already been indicated. It consists of a preliminary 
clarification of habits of thought about the motor carrier industry. 
For many different reasons, practically all parties directly at interest 
in motor carrier matters have tended to exaggerate the economic im- 
portance of regulation. So the first thing that needs to be done is to 
place motor carrier regulation in perspective. This process would in- 
evitably bring out certain facts: (1) The impact of motor carrier reg- 
ulation on railroad finances has diminished in recent years, and is now 
relatively small. (2) The impact of motor carrier regulation on the 
development of motor carriage in this country has been, and is, signif- 
icant. (3) Even this impact is probably a good deal less, in practice, 
than a study of legal texts and LC.C. decisions would lead one to be- 
lieve, due to the great number of important safety valves available to 
both shippers and carriers to reduce the theoretical pressure of the 
regulatory process. (4) Therefore the main problem with motor car- 
rier regulation is not its costs, which economists tend to exaggerate, 
but its absence of benefits. 

Step Two consists of answering the question : does regulatory mod- 
ernization with respect to motor carriage of freight necessarily re- 
quire complete and immediate deregulation ? There is really no way to 
provide a conclusive answer to this question by any known means of 
economic analysis; as a policy matter, a quasi-answer would be, "Not 

Step Three consists of an attempt to answer the further question : 
if motor carrier regulation were to be modernized, where, how, and 
with what objectives would one start? 

Given our restrictive system of regulating, not entry into the indus- 
try, but entry into particular phases of the industry, the big global 
problem of deregulation shatters into thousands of little local or re- 
gional or commodity or route or gateway problems of deregulation. 
One way to begin the process of regulatory modernization with the 
maximum prospect of obtaining the benefits of competition and the 
minimum prospect of leaving isolated shippers or consignees with- 
out common carrier service would be to free, from all regulation of 
rates, entry, etc., all routes now served by more than a specified num- 
ber of common carriers or general commodities. It is unfortunately 
true that neither the Interstate Commerce Commission nor anyone 
else can furnish a map of the United States showing such routes, but 
this is no answer at all. Without such detailed knowledge, sensible 
regulatory policy with respect to entry must in any case have been 
impossible short of expensive and time-consuming ad hoc investigation 
every time there is a request for a certificate of convenience and neces- 
sity. Moreover, this method of relaxing price controls and entry re- 


quirements would be quite in conformity with the "market dominance'' 
provisions of the Act of 1976 which apply to railroad rate-making 
(section 202 (b) and (c), amending Section 1(5) of the Interstate 
Commerce Act). The railroad legislation assures that there may be 
more competitive behavior in rate-making where some degree of com- 
petition is in fact pOvSsible. The motor carrier proposal suggested above 
would assure that there will be competitive rate-making (and pro- 
vision of service). 

The second thing that has to be faced is that reform of motor carrier 
regulation which would cut down on empty back hauls and circuitous 
routing is likely to meet less opposition if it enhances the relative 
position of tlie common carrier than if it reduces the economic role of 
the common carrier rehnftive to carriage of exempt commodities or 
private carriage. Therefore tlie most promising direction for regula- 
tory reform would seem to lie in permitting common carriers more 
flexibility with respect to arrangements for filling back hauls. Suoh a 
provision would be bound to create a host of temporary problems, 
since regulation has produced highly-artificial situations where com- 
mon carriers of the same commodities have back hauls in opposite 
directions. But this artificial situation is precisely what has to be 
eliminated. The most sti'aight forward way to eliminate it would be 
to grant more power to common carriers. 

A tliird issue that has to be faced, if not solved, is the problem of 
the large carrier with extensive operating rights who enjoys maximum 
opportunity to pick and choose traffic, to provide single-carrier serv- 
ice to shippers, and to combine the service a]^peal of relatively ra]^id 
delivery with the low costs which result from high load-factor opera- 
tion. In general, the largest trucking concerns have the most exten- 
sive operating rights. Therefore they can in principle be the bene- 
ficiaries of systems advantages which' are not adequately described in 
terms of costs alone, or scale alone. One way to introduce more com- 
petitive pressures into the industry without creating a free-for-all 
would be to permit end-to-end extensions of operatinir rijihts by 
carrier merger without the necessity for T.C.C. approval, plus expan- 
sion of.operating rights by existing regular-route common carriers at 
the rate of a given percentage increment per vear, a^rain without the 
necessity of I.C.C. approval. 

^ As for rates : the problem with regulation mav not be simplv (or en- 
tirely) that minimum motor carrier rates which are effective must be 
too high to clear the market, but that anv rate regulation of motor car- 
riers tends to bind an inherently flexible industry in a straitiacket— 
and a straitjacket. moreover, orio-inallv designed for railroad occu- 
pancy. This strait jacket is clearly less "binding, if binding at all. on 
precisely those rates which may still afl'ect motor-rail competition to 
at least a minor deorree. On the other hand, rates on less-than-truckload 
shipments doubtless have substantial importance v'?fh;n the motor 
earner industry not only as a means of enhancinir the value of operat- 
ing rights, but also as a levelling device to narrow the spectrum of 
ftlZ'Zfr ^7.^^"^^^^^V '""''''''^ ^^ ^Inppers. The exceptional prof- 
onernHnf '• 'J^''''' ^^^}^JF^ motor carriers with the most extensive 

nance .1 '7 f ""-'"''^'^ V^^^'"^' ^^'""^ ^^^ ^^™^ ^^ ""^'^^'^^'^t domi- 
nance are developing as far as possible from the bulk commodities on 


which railroad "market dominance-' is supposedly based. These excep- 
tional profits might well provide the rationale for applying at least 
some of the new pricing provisions of the 4-K Act to the motor carrier 

****** ^: 

Since this list does not pretend to be exhaustive, it can be ended, 
abruptly, right here. But it is worth noting that the approach im- 
plied by both suggestions two or three is designed to strengthen the 
position of common carriers as a group, relative to other types of 
carrier. Even suggestion one could be slanted in the same direction by 
restricting freedom of entry on "competitive routes" to carriers al- 
ready possessing operating rights which coyer part of the route or 
connect with it. The point to this whole approach is that, despite the 
manifest abstract claims of social justice which can be made on behalf 
of those who are not now able freely to enter the common-carrier 
trucking industry, concrete political obstacles can probably be over- 
come much more easily and with approximately the same final eco- 
nomic results if freedom of further entry is restricted to firms which 
are already in the regulated trucking industry. 


(By Theodore E. Keeler, 

University of California, Berkeley) 


This chapter provides an economic analysis of trunk airline regula- 
tion as practiced since 1938 by the Civil Aeronautics Board (CAB). 
Consideration is given to the' development and meanincr of the laws 
on which CAB regulation is based, the interpretation and execution of 
these laws by the CAB, and the effects of CAB behavior on fares, serv- 
ice quality, and producer returns in the trunk airline industrv\ 

The results indicate that in the years immediately preceding the 
CAB-s current drive to loosen its regulatory grip on airline fares and 
route entry, the welfare loss from its policies was high. ^lore spe- 
cifically, this and previous studies have found that between 1969 and 
1974, the CAB's policies cost trunk air passengers between $1 billion 
and $3 billion per year in fares, relative to what they would have paid 
if competitive market forces, rather than CAB regulations, were al- 
lowed to determine trunk air fares. These higher fares have not, how- 
ever, resulted in excess profits for the trunk airline industry. Since the 
advent of CAB regulation, the industry has earned a rate of return on 
total investment no greater than that earned in the corporate sector as 
a whole. The excess profits from the high CAB-level fares have in- 
stead been competed away through service rivalr\' bv the airlines. 
That is, rather than competinof through prices, the airlines have com- 
peted away potential profits by improving flight frequency and pro- 
viding more seating capacity for the number of passengers traveling 
than would exist in competitive markets. It would thus appear that 
CAB regulation has conferred benefits to passengers in the form of 
superior service quality, i.e., more frequent flights and easier reserva- 
tions during peak periods. To the extent that such benefits have ac- 
crued to airline passengers, however, there is also evidence that they 
accrued mainly to first-class passengers at least over the 1969-T-i 

The results presented here are supported not only by various eco- 
nomic models developed and estimated by the present writer and 
others (and described in the chapter), but also by more direct evi- 
dence from intrastate routes in California and Texas. These routes 
are beyond the regulatory grasp of the CAB. and on them coach fares 
have generally been no more than 40 to 66 percent of fares on equivalent 
CAB routes. Furthermore, flight frequencv on these intrastate routes 
would seem to be at least as great as flight frequency on equiA'alent 
CAB routes, and the intrastate (non-CAB-controUed) carriers flying 
these routes exclusively have tended consistently to earn profits at least 
as high as CAB -regulated carriers. 

In addition to providing empirical evidence on the effects of CAB 
regulation on fares, this studv analvzes in some detail the theoretical 
likelihood that the trunk airline industry will function efficiently and 
stably if competitive forces are allowed to take their course (that is, 
if current restrictions on firm entr\' and fares are eased). The results 



indicate that the trunk airline industry is indeed workably competi- 
tive. That is, if markets are allowed to function freely, the indus- 
try will function with at least as much efficiency and stability as most 
currently unregulated industries in manufacturing and wholesale and 
retail trade, and it T\'ill function considerably more efficiently than it 
has under CAB regulation. The only regulations needed to assure 
these results relate to considerations outside the current purview of 
the CAB, namely safety (airworthiness of pilots and aircraft), plus 
aircraft noise and emissions. Kegulation in these areas should 

In order to achieve the benefits of competitive markets in the airline 
industry, it is important to proceed gradually. Thus, this study sup- 
ports legislative proposals which guarantee the gradual expansion of 
authority by existing certificated carriers to new routes of their own 
choosing (the supplemental or charter carriers should be allowed to 
participate in this by expanding into scheduled service). Gradually, 
competition in the industry should be allowed to develop to the point 
whore entry to a route is determined not by regulatory fiat, but rather 
by the fitness, willingness, and ability of a carrier (new or already 
existing) to provide service on a route (based on safety and financial 
considerations alone). At this point, total deregulation of trunk fares 
should be allowed. 

It is quite unlikely that a meaningful number of communities will 
lose airline service as a result of such deregulation, for there is no 
evidence that current trunk airline profits from well-traveled routes 
are currently supporting service to smaller communities or that 
service to smaller communities by trunk carriers loses money. Never- 
theless, it is recommended that reform legislation should guarantee 
Federal support for service to all communities currently receiving it, 
for a period of 5 to 10 years. From evidence available, the cost of such 
support should be quite low. 

Finally, because under the proposed changes the airline industry 
would be treated like other unregulated industries in terms of firm 
entry and prices, it should also be treated equivalently in terms of 
antitrust. Thus, it is recommended that the airlines should be subject 
to the same antitrust laws (regarding mergers and restraint of trade) 
as other unregulated industries. 


(By Theodore E. Keeler) 

In recent years, the domestic trunk airline industry has been the sub- 
ject of numerous economic studies, nearly all of which have questioned 
the appropriateness of reofulation of this industry, as practiced by the 
Civil Aeronautics Board (CAB).^ Largely as a result of these studies, 
there have been increasing pressures for regulatory reform in airlines 
and le^rislation to achieve such reform has been proposed by members 
of both the executive and legislative branches of the L'^.S. Government, 
and even by a special task force of the CAB itself .^ 

In addition, since 1975. it has pursued a policv of deregulation based 
on the existing laws starting in the middle of 1977. 

It is the aim of thi^: chapter to ]u-ovKie an economic as-essment of 
the desirability of airline regulation as it is cui-rently practiced by the 
CAB. and to "evaluate the alternatives for regulatory reform. 

Scope of the Present Sti'dy 

Obviously, there is not space here to cover all aspects and effects of 
CAB regulation. While none are unimportant, some are more impor- 
tant than othPT'S. It is agreed by more or less all the above-mentioned 
studies that the most important effects of CAB policies are on sched- 
uled trunk passenger service (the definition of which is delineated 
in detail in the next section). It is this area, and periph.erallv the con- 
nected area of local service to small communities, on which the present 
studv will concentrate. 

The questions addi-essed here are straiirht forward, if the answers are 
not. AVIiat weie the conditions which led to the CAB regulation of 

^A partial list of such studies is Richard E. Pavps. .Air Transport and Ttc Rppu'itors 
(Oambridcp. Harvard T'nivprsitv Pross. 10r.2) : Mirhaol E. T.ovino. "Is RpMilntlon Xpces- 
sarv : ralifr.rnia Air Transnnrtatfon and National Rpculatnrv Poliry." YaIp T.nw Journ-l TH 
(JiiU- 190."). pp. 141fi-1447: WilliaTn Jordan. Alrllnp Rpsrulation in Ampriran : Fflfpcts 
and T'lipprfpptions fnaltimorp. .Tnhii« Hopkins U/iivprsity Prpss. 1070> : Thpodorp E. Kpplpr. 
"Airlinp Rpirnlatinn and MarkPt Pprformanrp." BpII .Tonrnal of Economirc 3 CAntnmn 
1071). Tip. .^09-424: Miohap] Pust.iv. Thp Efferts of Rpsrnl.ition on Rpsonr'^p Allorat'on 
in thp Donipstip Trunk Airlinp Tndnetry. Ph. D. dissprtation. Yale Univprsity. 197.1 : Gporsrp 
Doufrln'-- -ind .Tames C. Millpr NT. Economic Rejrulation of Dompstic Air Tnnsoor^ (W;ish- 
injrton. Tlip Brookincs Institution. 1974) : and Opor^p Eads. •Tompptition in thp Donipstic 
Trunk Airlinp Industrv : Too Much or Too Llttlp?" In A. Phillip's, pd.. Pr^motinir Com- 
pptit'on in Rpcrnlatpd MarkPts f Washington. The Rrookinirs In>;titut1on. 197.'). pv. l.'?-r.4. 

" Thpsp proposals stpm from studips such as R. Pulsifpr. L. Kpvs. P. Eldridge. J. A. 
^ff."Nr-^hon and W. L. Dpmory. Rpcruiatorv Rpform : Report of thp CAR Sppcial Sitaff 
fWa'-h-'neton. T'.S. Civil Apronautics Board. 197.")) : F.S. Spnatp. Committpp on thp .Tudi- 
ciarv. Snbconiniittpp on Administrativp Praoticps and Procpdures. Civil Aeron-'u^^ics P.nard 
Practices and Procpdurps fWnshinarton. T'.S. Govprnmpnt Printinsr Officp 19".")). rp^'prrpd 
to hprpaftpr as tlip Kpnnpdy Rpport (writtpn by S. Brpypr and associatps) : Simat. Hellip- 
«pn. and F'chner. An Analvsis of thp Tntr-^^tafe Air Carrlpr Rpsrulntory Forum : VoImttips 
T nnd II cXpw York, suhmittpd to thp F.S. Dpnartmpnt of Transportation. .Tanuarv 1970) : 
William T. Tolpmr^n Statpmpnt to the Aviation Suhcommlttpe on Commprcp Rpcrardinj: the 
Aviation Act of 197.T rWa^hincton. I'.S Government Prlntin<r Offlre. 1976). U.S. Seriate. 
Co"irvttpp on Oornpiprc. Scioncp. and Transportation. "Rpport on Ampudine thp Fpderal 
Aviation Act of 19."S (Washington. U.S. Government Printing Office. February 6, 197S). 



domestic airlines ? What are the ostensible reasons for this regulation, 
as set forth in the enabling legislation? How do these reasons relate to 
the goals set forth in this legislation, and to economic goals of efficient 
pricino; and resource allocation and a rapid rate of technological in- 
novation?^ How has the CAB gone about implementing these goals, 
Avhat are its policies today, and how do these policies relate to the orig- 
inal goals? What are the benefits and costs of this regiilation? To what 
extent are current policies inconsistent with economic efficiency ?^ To 
Avhat extent have social, economic, and technolo<rical conditions 
changed so as to render obsolete the original goals of CAB regulation? 
Finally, what, if any, reforms might be called for so as to improve the 
economic efficiency of the trunk airline industry without serious ad- 
verse effects on other social goals ? 

These questions will be dealt with in six broad sections. The second 
section presents a brief descriptive view of the industry, i.e., the vari- 
ous types of carriers and the functions they fulfill. The third section 
provides a brief historical sketch of the conditions which led to the 
Civil Aeronautics Act of 1938, and of the goals set forth in that 
legislation. The fourth section summarizes the specific regulatory pol- 
icies which the CAB has pursued in attempting to reach the goals of 
the Civil Aeronautics Act of 1938, as amended in 1940 and 1958. The 
fifth section, the heart and most important part of the study, analyzes 
the impact of CAB regulation in the framework of an economic model. 
This section contains not only a critical survey of previous work in 
this area, but also some new estimates of the welfare cost of CAB 
regulation, based on more recent data and somewhat different methods 
from previous studies. The sixth section deals with the overall appro- 
priateness of existing regulation and the desirability of change in 
principle. The final section sets forth the alternatives for policy 
change, and evaluates them in the light of previously discussed goals. 

One other dimension of scope of this study must be mentioned here — 
the dimension of time period covered. This report was originally writ- 
ten between May 1976 and February 1977. However, a year's delay in 
publication required an updating in the Spring of 1978. In the area 
of airline deregulation, much has happened during that period, and 
a complete updating of every detail of this chapter is impossible. How- 
ever, the policy analysis of legislative alternatives, given in the final 
chapter, has been updated so as to emphasize regulatory reform pro- 
posals dealt with by the second session of the 95th Congress in early 
1978. As relates to quantitative analysis of the effects of CAB regula- 
tion, while many of the facts and figures have been updated, consid- 
erable emphasis is still given to evidence for the years 1974 and earlier. 
The reason for this is quite simple : starting in 1975, the CAB has been 
pursuing a gradual policy of deregulation. Although this process has 
accelerated sharply with the appointment of Alfred E. Kahn, Kahn 
himself has pointed out that the process started with his predecessor, 
John Robson.^ Thus, starting in 1975, the CAB began to loosen its grip 
on airline fares, and figures for that year and later do not fully reflect 
the full effects of CAB regulation as it was practiced from 1937 to 
1974. (Of course, it might be also argued that legislative deregulation 
is unnecessary if the CAB has achieved what it has without legislative 

3 See interview by Barron's, reprinted in the Congressional Record — Senate (February 
21, 1978), p. S 1980. 


reform. The problem with that, as will be discussed further below, is 
that without legislative reform, or the pressure provided by legislative 
proposals, the CAB could eventually go back to its pre-19T5 policies.) 
The point here is that the years before 1975 give the best indication of 
what CAB regulation did"^with no threat of reform, and analysis of 
these years should provide the best basis for evaluation of the desir- 
ability of reform. 

TiiE Airline Industry 

Domestic airline service is provided bv several types of carriers, each 
with a different specialized function.^ The major carrier types and the 
traffic they account for are summarized in Table 1. 

As can be seen from Table 1. the domestic tnmk carriers account for 
tlie overwhelming majority of both passenger-miles and revenues. 
They are the larger airlines, serving liigh-density. non-subsidized 
7ou,tes in both domestic and international service. While domestic 
interstate routes are controlled exclusively by the CAB, this agency 
shares its regulatory powers on international routes with the Inter- 
national Air Transport Association, the State Department, the Office 
of the President, and, of course, the relevant regulatory agencies of the 
foreign countries seiwed. 

The CAB also maintains control over the local service carriers, 
nine airlines whose original justification was to serve smaller cities and 
towns across the country, and to serve as feeders to the tnmk system. 
The local service carriers have the right to Federal subsidies, to be 
allocated and distributed by the CAB (although the total amoimt of 
the subsidy depends on congressional appropriations). Very similar in 
function to the local service carriers are the Alaskan and Hawaiian 
carriers, who also receive Federal subsidies for serving the two non- 
contiguous States. 


passenger Operating 

miles Percent of revenues Percent of 

Carrier type (millions) total (millions) total 

Trunk 121,314 87.8 $9,549 82.1 

Local service -.. 11,028 8.0 1,200 10.3 

Intrastate 2,697 2.0 166 1.4 

Supplemental 1,805 1.3 2 429 3.7 

Alaskan 646 .5 94 .8 

Intra-Hawaii 645 .5 86 .7 

Cargo 4 (3) 97 .8 

Helicopter 24 (3) U .1 

Total 138,163 100.0 11,632 100.0 

' Excludes commuter carriers, for which revenue data are not available. 
» Includes revenues for international operations. 

I Negligible 

Sources: U.S. Civil Aeronautics Board, Air Carrier Traffic Statistics (December 1975) and Air Carrier Financial Statistics 
(Washington, U.S. Government Printing Office, 1976), and Simat, Helliesen, and Eichner (1976), pp. 11-81, 11-82, 111-15, 
and 111-19. 

Finally, the CAB regulates cargo and supplemental (charter) car- 
riers, but neither of these carrier types is allowed to provide sched- 
uled passenger service. 

* A more (iptalled doscription of the industry may be found in several of the above- 
mentioned reforences ; one of the most recent complete descriptions may be found in 
Douglas and Miller, pp. 192-197. 


In addition to passenger fares and subsidies, the CAB has control 
over charter rates, freight rates, and airmail contract costs paid by the 
Postal Service. 

Not all domestic carriers are regulated by the CAB, however — two 
important groups are not : the commuter carriers and the intrastate 
carriers. The commuter carriers provide low-density, short-haul serv- 
ice (both scheduled and taxi) with small aircraft (capacity under 30 
passengers and 7,500 pounds net takeoff weight). The CAB has the- 
power to regulate commuter carriers, but has granted them an exemp- 
tion. The intrastate carriers, currently operating in Texas, California, 
and Florida, provide short-haul, high-density services at fares below 
CAB levels. These carriers are of interest because they tell us some- 
thing about what airline service and fares might be like without CAB' 
regulation and/or subsidies. 

Because the trunk carriers account for such a large part of totali 
domestic airline service (87.8 percent of passenger-miles and 82.1 
percent of revenues), the present study will concentrate on them. The- 
story of how they came to be regulated is briefly presented in the- 
next section. 

The Development of Airline Regulation — A Brief Historical. 


Air transportation in the United States has had close Federal in- 
volvement since its very beginning.^ From the beginning at the end 
of World War I to about 1930, airmail was the only significant use 
to which air transport was put. Airmail service was supported by the 
Federal Government, first being provided directly by the Army, and, 
after 1925, being provided by private contractors supported by subsi- 
dies from the Post Office. 

By the late 1920's, however, the situation was changing. More spe- 
cifically, the Postmaster General appointed by Herbert Hoover, W. F. 
Brown, was a strong promoter of the airline industry, not only for 
the transport of mail, but also for the carriage of passengers.® Brown' 
lobbied strongly, both within the Hoover Administration and in Con- 
gress, for new legislation which would provide for the support of the 
airline industry, not only to carry mail, but also to promote its growth 
for other purposes, the carriage of passengers and freight. He got 
it in the form of the Watres Act of 1930, which required that the 
carriers act in compliance with "all rules, regulations, and orders that 
[might] be issued b}^ the Postmaster General for meeting the needs 
of the postal service and adjusting mail operations to the advances in 
the art of flying and passenger transportation." ^ In short, the Post 
Office was now placed in the position of promoting the airline indus- 
try for carrying passengers, as well as contracting with it to carry 
mail. Accordingly, this act based subsidies on plane-miles flown, 
rather than mail actually carried. Apparently, the most important 
reason for this shift was that air transport was believed to be an 

s More detailed discussion of reeulation durin? tlie earliest years of postal service may 
be foiind in Michael E. Levine. "Refrulatin? Airmail Transportation." Journal of Law 
and Economics 18 (October 1975). pp. 817-359, and P. T. David. The Economics of Air 
Mail Transportation (Washington, The Brookinss Institution. 1934). 

« Levine. "Regulating Airmail Transportation," pp. 320-321. 

'Ibid., p. 320. 


"'infnnt indiistiy," worthy of assistance to get it started (thus the 
Inference to "advances in the art of flying and passenger transpor- 

Tlie Postmaster General moved quickly to implement this law, ex- 
panding routes and improving service. However, he appeared to have 
<:'onsiderabl3' over-stepped the legal bounds of his authority in doing 
this : supposedly competitive bids were rigged so as to .oTi^rantee given 
carriers certain routes.^ Side payments were allowed when conflicts 
developed. A large portion of existing trunk airline authority was 
awarded to predecessors of the current trunk carriers during these 
spoils sessions. Not unexpectedly, these corrupt practices generated a 
scandal and cries for reform. 

Reform came with the Air Mail Act of 10.34. which placed regu- 
lation of fares and firm entry under the authority of the Interstate 
'Commerce Commission. This act provided for competitive bidding 
for route authority, as before, alonir with provisions to assure that the 
bidding would indeed be competitive. Once a carrier had won a route, 
if its awarded subsidies did not cover costs, it could petition the Inter- 
state Commerce Commission for an increase in subsidies (or rates) 
after 1 year of operation. 

Although competitive bidding for routes did occur as a result of the 
1934 Act, no routes changed hands. To assure that it would keep its 
existing routes, each carrier put in very low bids (often less than 1 
cent per route-mile). Carriers expected that once awarded routes, they 
would be granted rate increases before bankruptcy ensued. The law 
was somewhat vaffiie as to the basis for rate increases, however, and 
the ICC, uncertain as to the meaning of the law, was reluctant to 
grant increases. Threatened with insolvency, the airlines pressed for 
legislative reform. These pressures resulted in the Civil Aeronautics 
Act of 1938, which, largely unchanged to this day, serves as the basis 
for current airline regulation. 

Before considering the Civil Aeronautics Act of 1938, it is useful 
to digress briefly on the events which led to it. It is to this day asserted 
by some that the experience of the 1930's is a clear indication that in 
the absence of regulation, destructive competition would prevail in the 
airline industry. However, as has been pointed out in previous 
writing,^ the "destructive competition'' which occurred in the 1930's 
is the result of circumstances which have long ceased to exist. At that 
time, the industry was not economically viable without Federal subsi- 
dies. And the specific bidding scheme for routes embodied in the Air 
Mail Act of 1934 encouraged bids below costs because it held out the 
possibility of vSubsidy increases once the contracts were awarded. When 
these increased subsidies were not forthcoming, the carriers were the 
losers. This "destructive competition." then, is the exclusive fault of 
poorly designed subsidy legislation, and has nothing to do with what 
would happen to an airline industry which is viable and able to sup- 
port itself in the free marketplace. Nevertheless, as we shall see 
shortly, the Civil Aeronautics Act of 1938 expresses a concern for 
avoiding destructive competition. 

^ Another aspect of the background for the 1938 Act is worth con- 
sidering, before going on to the act itself. That is the "depression" 

Mbld.. pp. 321-322. 
" Caves, pp. 382-3S7. 


psychology which affected much of the economic thinking and social 
planning of the time : the Great Depression was associated with high 
unemployment and falling prices, whereas, in many peoples] minds, 
prosperity was associated with rising prices. Many economists and 
planners (especially in the Koosevelt administration) felt that if 
wages and prices for all groups could be rigged at higher levels, pros- 
perity would return. Furthermore, workers did not enjoy seeing their 
wages and prices fall, and legislation aimed at rigging them at arti- 
ficially high levels was politically attractive. This sort of thinking 
was responsible for many important laws and agencies, a number of 
which are still in existence, despite the fact that the economic theories 
behind them have been discredited, and the conditions have changed 
(examples include the National Industrial Eecovery Act of 1933, the 
Motor Carrier Act of 1935, the Wagner Labor Act of 1935, and the 
agriculture policies of Henry Wallace, which involved the waste of 
large amounts of agricultural produce in order to increase prices). 

The 1938 Act set up the Civil Aeronautics Authority (changed to 
the Civil Aeronautics Board in 1940), which had broad powers over 
the airlines. It was given complete control over entry of new firms, 
maximum and minimum rates (for passengers, air mail, and freight) , 
route structures of existing firms, and aviation safety. (Safety regula- 
tion was transferred to the Federal Aviation Agency in 1958 ; its name 
was changed to the Federal Aviation Administration in 1966.) What 
was the CAB to do with these powers ? Title I of the Act states : ^° 

In the exercise and performance of its powers and duties under this Act, the 
Board shall consider the following, among other things, as being in the public 
Interest, and in accordance with the public convenience and necessity : 

(a) The encouragement and development of an air-transportation system 
properly adapted to the present and future needs of the foreign and do- 
mestic commerce of the United States, of the Postal Service, and of the 
national defense. 

(b) The regulation of air transportation in such manner as to recognize 
and preserve the inherent advantages of, assure the highest degree of safety 
in. and foster sound economic conditions in such transportation, and to 
improve the relations between, and coordinate transportation by, air carriers ; 

(c) The promotion of adequate, economical, and efficient service by air 
carriers at reasonable charges, without unjust discriminations, undue prefer- 
ences or advantages, or unfair or destructive competitive practices; 

(d) Competition to the extent necessary to assure the sound development 
of an air-transportation system properly adapted to the needs of the foreign 
and domestic commerce of the United States, of the Postal Service, and of 
the National Defense : 

(e) The promotion of safety in air commerce : and, 

(/) The promotion, encouragement, and development of Civil Aeronautics. 

It is worth analyzing these provisions in terms of their relevance 
to the broad, general criteria for regulation set forth as a part of this 
project. They are only summarized brieflv here, because they are dis- 
cussed elsewhere in considerably more detail. ^^ Basically, economic 
theory suggests that direct government control of prices and/or firm 
entry is necessary when markets fail. Markets may fail under condi- 
tions of natural monopoly (where, because of scale economies, there 
is only "room" for one efficient firm in a market), externalities (where 
prices generated by the market do not correctly reflect costs to pro- 
ducers or benefits to consumers), or imperfect information (where 

i« Stat. 102, 72 Stat. 740. USCA 1.S08. 

n Caves, pp. 125-133 ; Douglas and Miller, pp. 197-205. 


producers are unable to seek out the highest price, consumers are 
unable to seek out the lowest price, or consumers are unable to deter- 
mine accurately just what product they are buying) . 

What references to market failure are included in these regulatory 
guidelines? There are a number of references in the legislation which 
would seem to suggest worries about market failure. 

The goal expressed in (c), that of "adequate, economical, and effi- 
cient service by air carriers at reasonable charges, without unjust 
discriminations, undue preferences or advantages, or unfair or de- 
structive competitive practices" is especially suggestive of economic 
aims. More specifically, the suggestion that regulation is necessary 
to promote "economical" and "efficient" service suggests that some sort 
of market failure such as monopoly may occur. The concern about 
price discrimination and destructive competition makes this clear. But 
it is appropriate to pause at this point and consider what these terms 
mean and why they represent market failure. This will be useful when, 
at a later time, we consider the appropriateness of current regulatory 
policies, as shaped by these provisions. 

Consider first the terms "economical and efficient." The term "eco- 
nomic efficiency" (which was more or less well-defined by economists 
in the 1930's as now)^^ referred to the notion that prices of all goods 
and services should reflect the extra costs of the resources used to 
produce an extra unit of a given good or service. This notion is com- 
monly referred to as marginal -cost pricing. In the case of the airline 
industrv\ economists have developed more elaborate notions of eco- 
nomic efficiency becnuse the service is not of uniform nualitv: rather, 
hicrh- and low-qualitv service are feasible, via variations in such 
dimensions at fliirht frequency, seating space, ease of getting a seat 
during peak periods, meals, and others. These more detailed notions 
of economic efficiency are discussed in more detail in section V of 
this chapter. The point here is that the act implicitlv assumes that 
relative to other industries, this industry requires regulation to achieve 
thic; efficiency, i.e.. it is subject to market failure. 

Further evidence of this is cr\ven in the reference to avoiding "unjust 
discriminations," for price discrimination, (i.e.. charirinrr different 
prices for goods of the same cost) will automatically be avoided when 
prices are set competitively.^^ The act's desire to avoid "unfair or 
destructive" competitive practices mnv be at some times inconsistent 
with the <roal of economic efficiency. There are two relevant economic 
concepts here. The first is that of predatory pricing, wherein one pro- 
ducer lowers his price below marginal cost with the aim of puttinir 
competitors out of business. Tt is only under relatively rare circum^ 
stances that predatory pricinrr represents rational behnvior for a firm, 
and it has very rarely been observed in any market. These points are 
discussed below. 

^-The notion that price should equal marginal cost to achieve economic efficiency and 
that mnrket? fail when they do not achieve mnreinnl cost pricinjr £roe<? hack nt least to 
the late lOth and enrlv 20th centuries in economic thought, and was pioneered hv the 
British economists Alfred :\rarshall and Alfred Pirrou : it cannot he said that all the 

hues wco worked out of this theory, however until the in40's and in'O'^ A fmod 
summary of all the issues Involved and of the development of economic thought on this 

l?E.oToVrt 7T?t!.l^i^^hsVp°;:Z%%'''''''"''' °' ""'"* ''"""^■" ^'""-'"■'>' ""'""" 

"In competitive markets, no one producer has enoufrh market power to charge customers 
different amounts, depending on their ahility to pay. With competition, there are enough 

Jf ho'/„T/^!i o"\''"-»?'■"'^."''^'; ^'"^ ""^ C^'^^^^ ^"* *o produce at the competitive price ; 
If he fails to do so, he will not sell any goods. , 


Destructive competition (sometimes lumped with predatory pricing, 
but treated here as a different concept) occurs when competitive 
market forces (rather than any desire for predation) drive prices 
below average costs for all firms in a market. "VYhen it happens, its 
effects on economic efficiency are often therapeutic. More specifically, 
destructive competition, as defined by economists, can occur when 
marginal cost (the cost of producing the last incremental unit of a 
good) is below average cost (simply total cost divided by the number 
of units produced). Under this situation, if pricing is non-collusive 
(i.e., competitive), price will be driven down to marginal cost, and 
everyone will lose money. This will continue until enough capacity 
leaves the industry to bring marginal cost back up to average cost. 
There are scA^eral circumstances under which this can occur : suppose, 
for example, that the industry is a natural monopoly, i.e., there is only 
enough "room" in the industry for one firm, because of scale econo- 
mies. Then, if several firms try to compete in the industry, since aver- 
age costs will be declining, marginal costs will be below average costs, 
and rate wars (i.e., destructive competition) will ensue until only one 
firm is left in the industry. One appropriate public policy in this 
case would be to allow one firm to take over the market, but to regu- 
late its rates so as to avoid monopoly pricing. Another possible situa- 
tion in which destructive competition can occur is in a declining 
industry, where there is excess capacity because of declining demand. 
Kate wars will again ensue until capacity has shrunk to match de- 
mand. This is a necessary and therapeutic mechanism for the achieve- 
ment of economic efficiency, for without it, capacity would not adjust 
to match demand. Finally, destructive competition can occur when 
there is temporary excess capacity becaiise of a business downturn. 
This is likelv to be a problem in an industry which has high fixed 
costs, but which faces large cyclical variations in demand (such as 
the cement industry). 

Even here, it can be shown that from the viewpoint of economic 
efficiency, markets should not fail, for such industries tend to earn 
exceptionally high profits during cyclical booms to offset losses during 
recessions.^'* At this point, however, it is important to remember how 
times have changed since the 1930's. There has been nothing since 
to compare with the severity and length of the cyclical downturn 
which prevailed then, and it is not surprising that legislation passed 
then should be concerned with destructive competition. The relevance 
of that legislation to the present period will be discussed lat^r. 

The provision for promotion of safety can be thought of as either 
an end in itself or yet another category of market failure. The market 
is quite capable of weeding out unsafe carriers because people would 
cease to ride them. But as the market worked itself out, many people 
could die in accidents, and that would be too high a price to pay for 
most citizens. We shall, in any event, be covering this issue in a subse- 
quent section^ (though safety is not a main focus of this paper) . 

The provisions in (a) are difficult to pigeonhole from an economic 
viewpoint. Even they, however, can b<^ given an economic reading-. The 
implication is that the free market will not generate "an air transporta- 
tion system properly adapted to the present and future needs of the 
-11 1 1 - 

" For a complete discussion of this point, see F. M. Scherer. IndustHal Market Struc- 
ture and Economic; Performance (Chicago, Rand McNally, 1970), pp. 198-206. 


foreign and domestic commerce of the United States, of the Postal 
Service, and of the National Defense.'- Externalities could be inferred 
here. The implication is that somehow, everyone in the United will 
benefit more from the existence of a complete, national airline system 
than do the citizenry from a private viewpoint. If this is correct, it 
makes sense to encourage the development of a larger (or different) 
system than the private market would generate. AVith indivisibilities 
and increasing returns to traffic density (i.e., something akin to a nat- 
ural monopoly) on a lower-density routes, this ma}' well be the case. 
The reason tliis is possible is that in a luxtural monopoly situation, scale 
economies are so great that marginal cost is below average cost. As we 
have alreadv stated, economic efficiency requires that price equal mar- 
ginal cost. To achieve an efficient solutiori in this situation, it is neces- 
sary to subsidize consumption of the good involved, making up for the 
difference between marginal cost and average cost. This will induce 
greater consumption of the good than would othei'wise occur. Whether 
or not such scale economies exist in the airline industry will be con- 
sidered later. 

Moreover, it may well be that a large network of airline routes with 
a high service level is useful for the national defense beyond what the 
market will provide during peacetime. But this issue nuist again be 
considered later on its own merits. 

There is some conflict in the goals listed in part (b), concerned as 
it is with ''preserving the inherent advantages of" and "fostering sound 
economic conditions in'' air transportation. "Preserving the inherent 
advantages of air transport implies that policies should encourage the 
use of air transport whenever and to the extent that its costs (relative 
to those of other models) justify it.'^ This can only be achieved with 
the marginal cost pricing mentioned above. On the other hand, any 
policies above and beyond this to "foster sound economic conditions" 
in the industiy could easily cause distortions between prices and mar- 
ginal costs. For example, rates set above the competitive level for the 
benefits of the industry will cause prices to be alx)ve marginal costs, 
and will fail to "preserve the inherent advantage" of the industry. It 
will cause underuse of air tra\'pl. On the other hand, a policy of sub- 
sidies (as provided by the Act and practiced by the CAB at the time) 
will cause overuse of aii* travel relative to the amount which would 
"preserve the inherent advantage" of the industry. At this point, it is 
important to remember that the airline industry in the lOoO's was not 
economically viable on its own (at least on the vast majority of routes). 
Without subsidies of some sort or other, it would die. Hence, "fostering 
sound economic conditions" in the industry would have been a neces- 
sity for a regulatory agency, even in the absence of a Great Depression, 
which made many normally profitable industries seem non-viable. 

Congress seems to have had these same considerations in mind in the 
specified goals of "competition to the extent necessary to assure the 
sound development of an air-transportation system properly adapted 
to the needs of the foreign and domestic commerce of the United 
States, of the Postal Service, and of the National Defense." Here, 

'^ The same notion of "preservinj? the inherent advantages of" and "fosterlnp: sound 
economic conditions in" each mode of transportation may be found also in the National 
Transportation Policy Statement of the Transportation Act of 1940. which established 
many of the same principles for surface transportation. The Economics of Transportation 
(Homevvood. 111., Irwin & Co., 1972), pp. 264-268, and the essay by J, R. Nelson in the 
present volume. 

83-944—78 7 


again, there appears to be a conflict between the desire to promote 
competition and the fear that competition will result in unsound eco- 
nomic conditions in the industry, rendering it incapable of fulfilling 
the goals of achieving a national system geared to the above-mentioned 
set of national needs. 

This conflict is evident not only in the law itself, but in the legisla- 
tive intent of the bill's f ramers. Thus, statements by the principal sup- 
porters of the bill in Congress at once expressed a fear of destructive 
competition, and a desire that if service on a given route were inade- 
quate, the Board should be required to admit new carriers.^^ 

Given the conditions which prevailed during the Great Depression, 
the reasons for this policy conflict and for the ambivalence which seems 
to be embodied in the law are easy enough to see. In discussing these 
conditions, it is important to start with an important premise : the in- 
dustry was not economically viable if left to market forces ; but on the 
other hand. Congress (and behind it, the public) had reasons to want 
to promote and preserve a national system : the postal service wanted 
to provide rapid airmail service over the entire nation; the military 
may have wanted a well-developed air transport system (with pilots, 
aircraft, airways, airports, etc., ready in case of war) ; and everyone 
may have felt that this was an infant industry, which, if promoted, 
would provide considerable benefits of high-speed transportation on a 
self-supporting basis at a later date. So, it is reasonable to infer a sim- 
ple intent in the Act of 1938: promoting and preserving a nation- 
wide airline system, despite the fact that this system was not at the 
time economically viable, i.e., would not be supported by the market. 
It is, of course, for this reason that the act elsewhere provides for sub- 
sidies, to be given out by the CAB."^^ 

Obviousiy, however, money to put into promoting air transport was 
and is limited in the quantit}^ available. Furthermore, it is possible to 
see how, with competition among airlines, there would be fear that this 
competition could eat up subsidy money needlessly : if two airlines were 
competing on a route, they could reduce fares in a continuous down- 
ward spiral, requesting more and more subsidies (indeed, this is some- 
thing like what happened in the airmail bidding of 1934). In the light 
of this, the provision for ''competition to the extent necessary" to pro- 
mote an adequate nationwide air transport system makes perfectly good 

Another potential conflict is hinted at by the "competition to the 
extent necessary" provision : in a subsidized industry, there is always 
a trade-off between the portions of costs recouped l3y prices and by 
subsidies. There is also a trade-off between the amount of service pro- 
vided (number of flights on a given route, and number of routes and 
communities served) and the amount of subsidy paid. The act does not 
specify explicitly how these conflicts are to be resolved, but it does give 
some hints : it provides that the air transport system promoted should 
bo national in scope, with the highest possible degree of safety, and 
adequate to the needs of the traveling and shipping public, the postal 
service, and the national defense. This must, however, be done in an 
"economical" way. When all these considerations are pulled together, 

i« A detailed discussion of the legislative Intent behind the Civil Aeronautics Act of 
J938 may be found in the Kennedy Report, pp. 208-215. 

1" The Act specifically entrusts the CAB with determining the "need" of a carrier for 
subsidies over and above "compensatory" air mail rates In Section 406. See Caves, p. 129. 


it can be argued that the Act requires the Board to promote a nation- 
wide system, adequate to all the groups mentioned above, at a minimum 
cost to the taxpayer. Yet a shorter way of putting this is to "minimize" 
subsidy costs subject to the constraint that an "adequate" amount of 
service be provided to a nationwide network of communities." 

While this sets forth the basis for CAB policies, there is obviously 
much room for interpretation still. Just how the CAB interpreted and 
executed the Civil Aeronautics Act of 1938 is the subject of the next 

CAB Policies ix Admixisterixg the Ci\t:l Aeroxautics Act of 


The 1938 act is, for practical purposes, unchanged to this day : the 
only near-impoilant changes occurred in 1940 when the CAB was re- 
organized to contain five persons, with 6-year terms, not more than 
three of which may be of one political party. In 1958, safety regula- 
tion was reorganized (given to the Federal Aviation Administra- 
tion), but economic regulation was kept the same. Naturally, much 
has happened in the nearly 40 years that this law has been in effect, 
and there is not space here to discuss it alL Some historical discussion 
of CAB policies is necessary, however, for an understanding of the 
present circumstances.^® 

policies toward xew firm entry 

Let us first review briefly the CAB's policies toward the entry of 
new firms. As regards trunk carriers (i.e., carriers specializing in high- 
density markets, with a large number of long hauls as well), the CAB 
has allowed no new firms to enter since 1938, despite the fact that be- 
tween 1950 and 1974 alone. 79 firms applied. ^'^ Ilowevor. the CAB 
has allowed entr^^ into markets in two ways: first, existing trunk 
carriers have been allowed new routes on numerous occasions. Second, 
just after AVorld War II, a new type of schedules air carrier came 
into being : the local service carrier. Initially, these lines were restricted 
to low-density, short-haul routes, supposedly not profitable and hence 
not attractive to trunks, given the larger aircraft used by the trunk 
carriers. For this., the local service carriers received subsidies. How- 
ever, in the late 1950's and through the 1960"s, the CAB allowed more 
and more local service carriers to serve trunk routes, on a limited, 
non-subsidized basis. In this sense, new firms were allowed to serve 
trunk routes. 

In another case, attempted entry by new firms had the same effects 
on airline prices as actual entry by new firms. This occurred in the late 
1940's and early 1950'?, with the advent of the non-scheduled carriers. 
These carriers, using war-trained pilots and war-surplus aircraft, 
attempted to provide transcontinental service in competition with 
the trunk carriers at drastically reduced fares, using much higher 
seating densities than did the trunk carriers (who. at the time, pro- 
vided only first-class service). The CAB ruled that the'^e carriers 

^"A detailed discussion of CAB policies up to 1061 mav be found in Caves, pp. 12.V2nO. 
and a summnry of policies from 1938-1974 may be found in U.S. Senate Subcommittee on 
Administrative Practice and Procedure, pp. 214-255. 

" Ibid., p. 6. 


were not to provide scheduled service, and hence the title irregular 
or nonsked, carriers. The trunk carriers responded (with the reluc- 
tant approval of the CAB) by providing limited amounts of sched- 
uled service with similarly high-density seating and low f ares.^^ This, 
along with CAB rejection of nonsked route applications, eliminated 
the threat of the nonskeds in domestic markets, while at the same 
time giving birth to coach service, which has grown to the point 
that coach travelers are now 90 percent of the total on trunk lines. 

The lesson of this story is that while the CAB has not allowed new 
firms to enter the trunk airline market in nearly 40 years, nevertheless, 
entry has not been barred so completely as to make one believe that 
perfectly-collusive price behavior would prevail in the absense of fare 
regulation both because of entry of existing firms on new routes 
( which has been permitted on a limited basis by the CAB ) , and be- 
cause of the nonskeds in the late 1940's. But the CAB does, of course, 
regulate fares, and it is to that topic which we now turn. 


"The remarkable thing about the regulation of fare levels by the 
Board is how little of it there has been." So Kichard Caves sums up 
the matter succinctly.^^ Caves backs this statement up with a persua- 
sive argument that, in fact, during its first twenty years of existence, 
the Board consistently attempted to dodge the issue whenever it was 
asked to come up with an overall, consistent policy regarding the level 
and structure of fares. It was not until 1961, with the General Passen- 
ger Fare Investigation, that the CAB proclaimed general principles 
of cost accounting (i.e., cost of capital and depreciation guidelines) 
on which fares should be based. And it still skirted general issues such 
as the load factor (per cent of seats filled) on which fares should be 
based, and the first-class-coach differential. It was not until the 
Domestic Passenger Fare Investigation, in the 1970's (discussed be- 
low) that the CAB met these issues head-on, and attempted to develop 
such general principles. 

But the mere fact that the CAB was for so long unwilling to openly 
articulate general policies regarding fares does not mean that it had 
none. It can be argued that the CAB preferred the flexibility provided 
by not having to apply general principles everywhere. Circumstances, 
change, and the CAB had every reason to want to change with them. 

Still, from the CAB's behavior over the years, certain dominant 
concerns can be traced. Consider first the pre- World War II period 
during the first years of Board regulation. It has been said that the 
Board and the carriers had no concern about fares at that time.^^ It 
might be more accurate to say that there was agreement between the 
Board and carriers as to what fares should be, and hence no conflicts 
arose. Both agreed on a discriminatory, "value of service," pricing 
scheme, based on the following premise. To promote the development 
of civil aviation, as required by law, it was necessary to have a nation- 
wide system, despite the fact that some routes were more profitable 

*^ Coach fares were not so low as the nonsked fares, but the service was more convenient, 
given that coach service operated on schedules and was hence more reliable. For a more 
detailed discussion of the nonskeds, see Caves, pp. 171-174, 370-371. 

^ Caves, p. 140. 

*« Ibid., p. 141. 


than others. To assure this, it was necessary that prices be set to moke 
the service attractive on all routes. Thus, during those days, airline 
fares were based almost exclusively on first-class rail fares (indeed, 
they were roucfhly the same in most major markets).-^ The prices of 
the nearest substitute determined fares, and government subsidies 
made up the rest. It is not difficult to see why the CAB and the car- 
riers should arrive at a tacit agreement as to the appropriateness of 
this scheme. 

During the first 10 years of CAB regulation, fare changes occurred 
only when it was thought that the carriers* returns were excessive or 
inadequate. Thus, fare cuts were required during TTorld War IT in the 
face of hicfh load factoids and large profits. Fare increases were allowed 
after the war as profits declined. 

But in the late 1040-8. this situation changed : pressures arose for 
fai-e cuts, generated by the rise of the aforementioned nonskeds. The 
CAB'S response to these pressures had a certain consistency to it : it 
wns pervaded by a fear that airline revenues and profit would decline. 
"Wlienever the requested price cut was to a limited class of customers 
and allowed clearcut discrimination, without loss of revenues (as in 
the case of family fares), the CAB allowed the cuts.-'* On the other 
hand, when the proposed reductions threatened to lower the entire fare 
structure, the Board was at best reluctant to go along (the CAB's 
slowness to accept coach fares at the Ix^ninninof and its reluctance to 
allow their spread is a case in point ).-^ This behavior can be criticized 
as anti-consumer. But it must be remembered that during the late 
1040's. the trunk industry was still subsidized directly, and only the 
benefit of hindsight allows us to see that direct subsidies would soon 
no longer be necessary. Tender those circumstances, the CAB's fears of 
increased subsidy needs because of the onset of price competition \fere 
not irrational. 

During the lO.^O's. however, the economic situation of the trunk 
carriers changed. They went off direct subsidies, and more or less all 
of their unprofitable routes were taken over by the local service car- 
riei-s. Yet still the CAB tended to prevent fare cuts, and to prevent 
entry by lines who propose^l such cuts. This was especially the case 
when carrier returns were below the official returns ''allowed'' by the 
CAB (discussed later in this section), or when a weak carrier was 

Rate of return regitlation 

What are the motivations for this? One most cei-tainly relates to a 
sincere desire that the airlines should be able to earn a ^'fair' rate of 
return. ^lore specifically, as a result of the General Passenger Fare 
Investigation of 1961, the CAB established that a "reasonable'' rate of 
return for the trunk carriei-s was 10.5 percent overall, including both 
debt and equity capital (i.e.. the return to both stockholders' equity 
and bondholders' debt). As a result of the Doinestic Passenger Fare 
Investigation of 1971, this overall return for the tmnk carriers was 
raised to 12 percent. 

»lbld.. p. 357. 

2* A discussion of price-discriminatory policies of the CAB may be found In CaTes, 
pp. 15.")-! 67. 

2^ Caves, pp. 167-168. 

r 90 

There is little wonder that there should be competitive pressures to 
reduce fares set to yield these rates of return, because they are con- 
siderably higher than the rate which could be earned by an equivalent 
investment in a typical U.S. corporation (this "typical" return is 
called a "normal" return, or "the opportunity cost of capital," because 
it gives an indication as to what other typical investment opportuni- 
ties will yield) . More specifically, over the 40-year period between 1938 
and 1968, the typical U.S. corporation earned an average return on 
investment of about 7.5 percent after taxes.^® Thus, the CAB, in the 
retui-ns which it has allowed over the past 15 years, may be building 
an incentive for price-cutting or overscheduling by setting them at 
"too high" a level. 

Despite the CAB's persistent aim of a rate of return of 10.5 to 12 
percent on investment for trunk airlines, the carriers have seldom been, 
that profitable, and on the average, they, like the corporate sector, have 
earned an average of about 7.5 percent on their investments during 
peacetime years between 1938 and 1966.-^ Thus, CAB policies have 
been unsuccessful at achieving the above-normal profits embodied in 
its regulatory goals, despite the persistent vigilance on the part of the 
CAB to prevent price competition. We shall consider the reasons for 
this failure shortly. First, however, we investigate another reason 
why the CAB has desired to prevent price competition, and the degree 
of success it has had on this second count. 


This second reason is the CAB's desire to minimize the need for 
Federal subsidies to unprofitable routes. The CAB has felt the political 
obligation to support service on these unprofitable routes for reasons 
described in the previous section. The idea here is that if some routes 
are profitable, the profits from these routes can be used to support losses 
on unprofitable routes (this is generally called "cross-subsidization" 
by economists). If, however, price competition is allowed on profitable 
routes, the profits will be dissipated, and subsidy requirements for un- 
profitable routes will increase (of course, competition in flight fre- 
quency may cause the same thing; this is discussed below.) 

A clear statement of the CAB's views on cross-subsidization came in 
the late 1940's, when "nonsked" carriers (previously mentioned) at- 
tempted to enter long-haul, high-density routes with scheduled service 
at fares considerably below those charged by the trunk carriers. The 
CAB rejected the nonskeds' proposals, for the explicit reason that such 
low fares would eliminate trunk profits from these routes, necessitating 
an increase in subsidies for unprofitable routes.^^ 

The CAB applied its cross-subsidization doctrine again in the early 
1950's when the local service carriers were new, and were expanding 
their routes. During these years, the CAB on several occasions with- 
drew the route authority of the trunk carriers from a given route, and 
gave it to a local service carrier, stating explicitly that if local service 
carriers had some profitable routes, it would greatly reduce govern- 
ment subsidy requirements. For example, in the Southwest Reneioal 

29 For returns up to 1968. see Caves, p. 392. For returns past 1962, see Theodore E. 
Keeler, Resource Allocation in Intercity Passenger Transportation (Doctoral dissertation, 
Massachusetts Institute of Technology, 1971), pp. 20, 22. 

2'^ See Keeler, Resource Allocation, pp. 20, 22, and Caves, p. 392. 

28 Caves, p. 173. 


case of 1952, the CAB suspended United's^ route authority (over 
United's strong protest) to four California cities and awarded them to 
a local service carrier. The Board noted that the local service carriers 
involved (directly and through connections) would gain an estimated 
$207,000 "which 'would ultimately revert to the Government in the 
form of reduced subsidy mail-pay requirements." ^^ Numerous other 
cases of this can also be found during the period in the early 1950's 
when the local service carriers were developing their route structures.^" 

In the middle and late 1950's, with strong Congressional support for 
higher subsidies, the CAB lost interest in route-strengthening, and it 
worried less about it.^^ However, by the early 1060's, the situation 
changed again. President Kennedy's 1962 transportation message to 
Congress expressed a desire for the CAB "to develop ... a step-by- 
step program, with specific annual targets, to assure sharp reduction 
of operating subsidies." ^^ The Board again embarked on a route- 
strengthening drive, wherein it granted the local service carriers routes 
competitive with trunk carriers. Thus, whereas in 1955 only 19 percent 
of local service passenger-miles came from routes with the rivalry of 
more than one carrier, by 1965 fully one-third of all local ser^dce pas- 
senger-miles came from such routes. Further route-strengthening was 
attempted between 1966 and 1972, and by 1970 fully 45 percent of all 
local service passenger-miles were generated on competitive routes.^' 

The success and the desirability of such attempted route-strength- 
ening will be discussed below. The point here is that in awarding 
routes so as to reduce the subsidy needs of the local service carriers, 
tlie CAB was clearly admitting that some routes were more profitable 
than others, and acknowledging that its policy was indeed to use 
profits from some routes to support losses from others. 

'Wliile the strongest evidence of desire to cross subsidize comes 
from the CAB's behavior toward the local service carriers, it has 
often behaved in a similar manner toward the trunks: a trunk carrier 
if often awarded a profitable new route if its existing route structure 
makes it unprofitable (after all, if a trunk carrier with improfitable 
routes went out of business, the CAB micfht have to come up with 
subsidies to cover operations of its routes by subsidized carriers) . An 
important example of this occurred in 1956. when the Board granted 
Xortheast Airlines entry into the Xew York-Miami market, with the 
explicit goal of "strengthening" its route structure. 

How successful have CAB policies on cross-subsidization been? 
Direct evidence on this count is difficult to ^xct. However, there are 
several pieces of evidence from which to infer that little cross-subsi- 
dization iioes on, at least at the trunk level. 

First, it is worth noting that as soon as the trunk carriers went off 
direct subsidies, they were generally more than willing to abandon 
their unprofitable routes, allowing them to be taken over by local 
service carriers. Why should their stockholders be eager to subsidize 
service to small towns from their own pockets ? ^* Then, as the local 

2» Pnves, pp. 224-225. 
a^Ihid.. pp. 224-229. 

31 Thid.. p. 220. 

32 Heorcre Earls. The Local Service Airline Experiment (Washington, The Brookings 
Institution. 1971). p. 107. 

"Eads. p. 118. 

5* A complete discussion of the development of the local service carriers, and of the 
abandonment of various routes by the trunks, may be found in Eads, Chapter 6. 


service carriers have been granted more and more profitable routes, 
they have been eager to go off subsidies, and to hand off their un- 
profitable routes to unregulated third-level or commuter carriers.^^ 
Thus, Allegheny has acquired enough profitable routes to go off sub- 
sidy completely, and has contracted out most of its unprofitable routes 
to third-level carriers using much smaller aircraft.^^ The point is that 
just as' rapidly as the CAB seems to grant profitable routes to cross- 
subsidize unprofitable ones, the carriers involved attempt to abandon 
the unprofitable ones. And in the past 6 years or so, the CAB has been 
willing to grant such abandonments freely. 


Distance (miles) 

Volume of traffic (passengers per year) 

200 and 




Over 1,200 

Under 1,000 






1,001 to 5,000 


5,001 to 25,000—. - - _ ... 


25,001 to 100,000 


Over 100,000 


Source: Caves (1962), p. 409. 

Second, Caves provides direct evidence that even in the late 1950's, 
cross subsidies had ceased to exist on trunk level routes. Table 2 shows 
Caves' calculations as to the relationship between price and cost by 
route density and length of haul as of 1958. His results indicate rather 
clearly that there was relatively little cross subsidy on routes above 
200 miles in length and above 5,000 passengers per year in density. As 
Caves points out, practically all trunk routes by 1958 per 200 miles or 
longer and were above 5,000 passengers per year in length. It can be 
argued, however, that with the advent of the jet plane, the cost of 
long-haul travel relative to short-haul travel declined, and CAB fares 
by long-hauls.^'' However, this was corrected more or less completely 
by fare changes in the early 1970's.^^ Finally, the CAB itself gave 
evidence of rejecting the cross-subsidy doctrine in 1974. It stated that 
where fares exceed costs "rather than providing subsidization of short- 
haul services, . . . [they] merely subsidize wasteful competitive prac- 
tices . . . Only if the domestic air transportation system consisted of 
a single monopoly carrier" could cross subsidy be expected to succeed.^^ 
(The "wasteful competitive practices" mentioned by the CAB will be 
discussed in the next section.) 

There may still be a small amount of cross-subsidization going on 
from high-density routes to low-density ones. More recent and de- 
tailed evidence will be considered later. But the objective evidence 
which exists would imply that such cross subsidies are infinitesimal 
relative to the total costs or revenues of the domestic airline industry, 
and affect at most only a very small proportion of passengers. 

35 Ibid., pp. 166-169. 

^ In 1974-1977, Allegheny received no subsidies. See Civil Aeronautics Board. Air 
Carrier Financial Statistics (December 1975), p. 23, (June 1977) p. 22. 

3'^ "Airline Rej;ulation and Market Performance," p. 419. 

38 The increase in fare rnarkup ovpr costs with distance observed for 1968 seemed to have 
disappeared by 1972. See Keeler (Ibid), p. 420. 

3» CAB order 74-382, pp. 71-72, as cited in Kennedy Report, p. 64. 


If cross subsidies are non-existent, and if the airlines have earned 
normal profits, one is tempted to believe that CAB regulation has had 
no effect on fares. And yet evidence from markets outside the reach 
of the CAB. the intrastate markets in Texas and California, would 
imply that this is not the case. Thus, intrastate fares in California, 
brought about by competition from non-CAB regulated carriers (most 
importantly Pacific South-west Airlines (USA)), have consistently 
been 30 to 50 percent below fares on equivalent interstate routes. Simi- 
larly, since the entry of an intrastate carrier in Texas, Southwest Air- 
lines, fares in that State have been lower by a similar amount.^^ More- 
over, potential entrants to interstate routes have proposed fares simi- 
larlv lower than CAB regulated fares on long-haul routes. Most 
notably. World Airways, a charter carrier, has proposed Xew York- 
California service at faVes 40 percent below prevailing CAB regulated 
fares. Finally, the CAB has itself authorized fares to Hawaii and 
Puerto Rico far lower than long-haul mainland fares for similar dis- 
tances, and the eagerness of carriers to provide this service without 
subsidy is evidence that this service is profitable.^^ ^More detailed evi- 
dence on these fares will be considered shortly. The main point here is 
that there is ample evidence that CAB policies have indeed rigged 
trunk air fares in continental United States at levels well above what 
an unregulated market would provide for. 


If, indeed, fares are rigged at such high levels, where do the money 
(and, more importantly, the accom])anying resources) go? We have 
already noted that they do not go into excess profit, and there is little 
evidence to support the notion that it goes into cross subsidizing low- 
density routes. 

This seeming paradox is not likely to be a quandary to anyone who 
flies, and the answer has been noted in numerous studies. As the present 
writer put it in 1972, "with faros set at high cartel levels, the airlines 
have competed away profits through excess capacity. Since more fre- 
quent flights mean more business, there is an incentive to compete 
through flight frequency instead of prices, and the resulting excess 
capacity appears to keep profits down to a normal level.'' ^- Of course, 
service quality competition can occur in dimensions other than flight 
frequency. Holding frequency constant, larger airci-aft with more seats 
make it easier to get reservations during peak periods: roomier seats, 
lounges, movies, sumptuous meals, and fast baggage service are other 
ways to compete in service qualit}^, as well.^^ Different students of the 
industry may disagree as to just how close the industry has come to 
earning normal profits, and to just what types of service quality com- 
petition have been used to compete profits away. But all economists 
studying the problem (and the Senate Subcommittee on Administra- 
tive Practices and Procedure) seem to agree that the airlines do com- 
pete away more or less all excess profits in service quality competi- 
tion, thereby negating CAB effoi-ts to raise trunk profits above a 

^ This is documented in detail in the following section. 
<i Kennedy Report, pp. 49-51. 

*- "Airline Regulation and Market Performance." p. 421. 

*3 Lower load factors make It easier to get reservations during crowded periods. This 
notion will be discussed in more detail In the following section. 


normal level, and to cross-subsidize service on lower-density lines. 
We sHall return to this matter in the next section. 


The CAB has for some time been aware of the service quality com- 
petition of the sort noted above, and has attempted to do something 
about it. More specifically, the Domestic Passenger Fare Investigation 
of 1971-74 (DPFI) was intended to develop (for the first time) an ail- 
around fare policy, and to deal with the service quality competition 
problem, and some others, as well.** Let us therefore briefly consider 
the DPFI, the policy changes resulting from it, and the actual changes 
in fares and service qualities which in turn resulted from these policy 

The first change in CAB goals resulting from the DPFI was to up 
the target rate of return to be built into fares from 10.5 to 12 percent 
rates of return (in both cases this is the after-tax return on all assets, 
debt and equity combined). The appropriateness of this change is 
highly questionable, especially given that the industry has managed to 
get all too much capital at a much lower (but normal) return of 7.5 
percent after taxes. 

In addition to a rate of return standard, the DPFI resulted in a load- 
factor standard of 55 percent overall, i.e., fares were to be set such that 
a carrier could earn exactly a 12 percent after-tax return on all assets 
if 55 percent of the seats were filled. This change, unlike the target 
rate of return, was based on a new principle. The CAB has previously 
set target rates of return, but it had not set a target load factor. Previ- 
ously, if the return was "inadequate," the Board could authorize fare 
increases ; but if additional service competition drove the return back 
down to an "inadequate level," another fare increase would be needed, 
and so on. This "rachet effect" could theoretically drive fares ever 
higher, with no increases in cost at a given load factor.*^ 

In principle, the notion of a 55 percent load factor standard on 
which to base fares was an improvement, relative to past practice, 
which was vulnerable to the aforementioned rachet effect. However, 
there are at least two reasons why the policies resulting from the 
DPFI may fall well short of what is feasible by way of economic 
efficiency (note we say "mxay" here, because a complete discussion of 
economic efficiency in the airline industry is deferred until the next 

First, the assumption of a 55 percent load factor is arbitrary, and 
there is little, if any, evidence that it represents an optimal trade off 
between fare and service quality (i.e., the ease of flying when one 
wants to fly) on a significant number of routes. In fact, evidence 
presented in the following section indicates that on most routes a 
higher load factor would be appropriate. Furthermore, it is likely 
that the optimal load factor will vary from route to route, making 
a single universal number inappropriate. 

^ A more detailed discussion of the DPFI may be found in Douglas and Miller, pp. 

*» Tlie term "rachet effect," applied to this phenomenon, was first used by Douglas and 
Miller. See pp. 54-57. 

*fl For a more detailed critique of the DPFI, see the Kennedy Report, pp. 121-122, and 
Coleman, pp. 6-13. 


Second, we have already presented evidence that a 12 percent rate 
of return is probably higher than the opportunity cost of capital, and 
higher than the airline industry has historically generated. On the 
basis of past evidence, one \vould expect rates set to generate a 12 
percent rate of return at a 55 percent load factor to, in fact, generate 
a lower post-tax return (say, 7.5 percent), as service quality com- 
petition drives the load factor overall below 55 percent.*^ That is,. 
load factors would be driven down until airline profits equaled the 
opportunity cost of capital. 

So far, we have been primarily concerned with describing the goals 
and policies of the Civil Aeronautics Board in its execution of the 
Civil Aeronautics Act of 1938, with only a hint here and there as to 
the economic appropriateness of its policies. It is now time to turn 
to a more systematic economic evaluation of these policies. 

The Impact of CAB Regulation: Ax Economic Analysis 

At the end of the previous section, I described a hypothesis about 
airline regulation on which most students of the area agree — that, 
although the CAB has set trunk airline fares at high, "cartel" levels, 
the potential profits' from tliese fares are competed away through 
frequency and service quality competition on the part of the airlines. 
As a result, tlie airlines do not gain profits from the regulation, and 
the consumer is left paying a fare much higher than he would prefer, 
with lower load factors and presumably higher service quality than 
he would prefer.^^ 

This section is concerned with examining the evidence regarding 
this hypothesis, and with providing an estimate as to the likely wel- 
fare loss or gain from CAB regulation. Consideration must be given 
to costs and benefits to several groups: consumers in high-density 
markets, consumers in low-density markets (wliich perhaps currently 
are cross subsidized), carriers, and suppliers of inputs to carriers. 
We shall postpone our discussion of the effects of various regulatory 
policies on broader considerations such as the national defense and 
the postal service to a later section. 


In order to test the hypothesis that CAB regulation has held fares 
at an unnaturally high level, and load factors at an unnaturally low 
level, the first place one would be inclined to look would be domestic 
markets outside the reach of the CAB's powers. Such markets cur- 
rently exist on intrastate routes in California and Texas. 

^larkets in both States have enjoyed considerably less regulation, 
both of fares and of firm entry, than have the intrastate trunk routes 
controlled by the CAB. Both are regulated to some degree, however, 
in California by the California Public T^tilities Commission, and 
in Texas by tlie Texas Aeronautics Commission. 

•*'' At no time In tho post 20 years have the airlines earned a return as hich as 12 percent 
overall. If the cost of capital of 12 percent assumed bv the CAB were correct, most firms 
would have {rone out of the Industry years ajro. See Kennedy Report, pp. 121-122. 

*^ The Idea here is that hipher service ouality comes from higher fares, and there is 
a limit to -what the typical traveler Is willing to pay for high service quality. This is 
discussed at length below. 


In California, both fares and firm entry were almost totally unreg- 
ulated until 1965, except that a new firm had to meet safety standards 
set by the Federal Aviation Administration, and the PUC could set 
maximum rates, but not minimum ones. In 1965, the Public Utilities 
Commission was given the right to control both fares and entry of 
firms, with the aim of achieving an ". . . orderly, efficient, economical, 
and healthy intrastate air network . . ." along with stability, low 
fares, and frequent service.^^ 

The Texas Aeronautics Commission (TAC) has complete control 
over entry of firms, as does the California PUC, but very limited 
control over fares. Technically, it has control over an intrastate 
carrier's fares; but it claims no control over the fares of intrastate 
operations of interstate carriers, so it has effectively allowed pricing 
freedom for both carrier types.^^ The law requires the TAC to "fur- 
ther the public interest and aeronautical progress by providing for 
the protection, promotion, and development of aeronautics." ^^ 


Fare Nonstop Fare 

(includes distance per 

tax) (milas) mile 

Los Angeles to San Diego 

Fresno to San Francisco 

Fresno to Los Angeles 

Los Angeles to San Francisco 

Los Angeles to Sacramento 

Cleveland to Pittsburgh 

New York to Hartford 

Baltimore to Norfolk 

Detroit to Dayton 

New York to Syracuse 

New York to Washington 

Chicago to Minneapolis 

Pittsburgh to New York 

Boston to Baltimore 

Los Angeles to Phoenix 

1 La Guardia to Washington National. See Simat, Helliesen, and Eicher, vol. I, exhibit 7. 
Source: Official Airline Guide, Feb. 1, 1977. 

In both States, the relevant regulatory agencies have been consider- 
ably more liberal than the CAB in allowing new firms to enter, and 
in allowing downward price competition. Thus, before 1965, 16 new 
carriers entered the California market, and after 1965 two new car- 
riers have been allowed to enter in California. Of these 18 carriers, 
however, only two important ones, Pacific Southwest Airlines (PSA) 
and Air California have survived.^^ In Texas, one carrier. Southwest 
Airlines, has entered the market, and then only after lengthy court 
litigations although the Texas Aeronautics Commission favored entry 
of Southwest Airlines from the beginning, Texas law allows court 

*» Simat, Helliesen, and Eichner, vol. I, p. 11. 

«» Testimony of Charles A. Murphy, Executive Director. Texas Aeronautics Commission, 
before U.S. Senate, Committee on the Judiciary, Subcommittee on Administrative Prac- 
tices and Procedure, Hearings on Oversight of Civil Aeronautics Board Practices and 
Procedures, Feb. 14, 1975. 94th Congress, First Session, rol. II (Washington, U.S. Govern- 
ment Printing Office, 1975), p. 528. 

5^ Simat, Helliesen, and Eichner, vol. I. p. 11. 

^2 It is worth noting, however, that the carriers which went out of business in California 
tended to be very small and short-lived (under a year, usually). Thus, this experience 
does not indicate that a well-established carrier would be likely to go out of business 
■with deregulation. See Jordan, pp. 14-33. 



$0. 112 












































appeal of TAC decisions by interested parties, including carriers al- 
ready serving the routes proposed for service by South west.^^ 

The numbers of intrastate carriers succeeding in each market (one 
in Texas and two in California) may seem small. But even these small 
numbers would seem to be enough to instigate vigorous price competi- 
tion. They cut fares sharply and forced the CAB-regulated carriers 
to match the cuts for intrastate traffic.^* As a result, fares on every 
intrastate route are significantly below fares on routes of equivalent 
length and traffic density as CAB routes. 

Fare differences 

To give an idea of these differences, Table o presents some current 
California intrastate fares (and yields per mile) alongside equivalent 
figures for routes of similar length on CAB-regulated intrastate 
routes.^^ The results are revealing. They indicate tliat fares on inter- 
state routes are consistently greater than fares on equivalent Califor- 
nia routes by 70 to 120 percent. Thus, the fare for the Pittsburgh- 
Cleveland route (105 miles) is $27, or over 120 percent greater than 
the Los Angeles-San Diego fare of $12.25 (for a trip of 109 miles). 
Similarly, the fare from Xew York (La Guardia) to Washington 
(National), a distance of 215 miles, is $'^>7, or 80 percent greater than 
the fare foi- Los Angeles-Fresno of $20 (for a trip of 218 miles). As 
the reader can verify, the results are more or less the same for compari- 
son of any of the other comparable city-pairs. 

Similar fare savings are available in Texas. Between Dallas and 
Houston, a distance of 222 miles, the curi'ent fare is $25 on weekdays 
before 7 p.m., and $15 for evenings and weekends (both figures include 
t?«xes).'^^ This compares with a fai'e of $o7 (including taxes) for the 
New York-Washington route, a distance of 215 miles (Table 3). 

The low fares provided by the intrastate carriers in California and 
Texas are not achieved at the cost of safety or moderjiity of aircraft. 
The intrastate carriers use aircraft identical to those of interstate car- 
riei-s on equivalent routes, and their safety records are excellent in 
comparison with CAB-certificated carriers.-^'' P'urthermore, it is not 
clear that the service quality provided by the intrastate carriers is 
overall inferior to that ]:)rovided in coach service on ec^uivalent routes 
by interstate carriers. Both genei'ally use coach seats 18 inches wide 
pitched 84 inches apart. Although the intrastate carriers do not serve 
meals, the trunk carriers do 7iot geneially serve meals or free snacks 
on the hops of 65 to 850 miles served by the intrastate carriers. As re- 
gards flight frequency, the evidence from a number of routes where 
new low-fare competition has been instituted by an intrastate carrier 
would suggest that the low fares charged by intrastate carriers will 
induce sufficient new demand to support more flights than the higher 
fares previously charged by CAB-certificated carrieis. Finally, the in- 

^■* Simat, Helliespn. and Eicliiipr. pp. 17-18. 

■"* A fnU disoussioii of this )iri(inc; behavior may be found in Jordan. Thapter 5 for 
California, and in Siniat, Ilelliesen. and Eichner, vol. II, chapters II and III for California 
and Texas, respectivelv. 

'■-''It was pointed out'at the hejrinninjr of this chapter that by 1077 the C.\B had already 
embarked on a pro^'rani of very modest derepnlation ; this means a 1!>77 comparison mijrht 
not show the full effect of derc-jrulatlon of CAB routes, relative, say, to 1974. However, these 
comparisons are for repulnr fares, whereas in early 1977. it was mainly in discount fares 
th.Tt the trunk carriers made reductions. 

«> Dallas-Houston fares come from the Official .Virline Guide. Feb. 1. 1977. p. 451. 

"'Testimony of J. Barnum, Hearings on Orcrsight of Civil Aeronautics Board Proxtieen 
and Procedures, vol. I, p. 10. 


trastate carriers have made considerable use of "satellite" airports, 
providing service to and from more points in a metropolitan area than 
generally occurs on interstate routes, and making for a higher level of 
passenger convenience. It is thus not obvious that intrastate passen- 
gers have inferior service compared to interstate coach passengers. 

Causes of lower intrastate fares 

How do the intrastate carriers achieve these lower fares? It is 
certainly not by accepting lower profits than the CAB-certificated 
carriers. Once start-up costs were covered, intrastate carriers have con- 
sistently enjoyed profits as great or greater than those of CAB-certifi- 
cated carriers. Thus, in 1972 through 1974, Air California earned a re- 
turn on over 24 percent after taxes on its equity investment.^^ Over 
the same period, PSA earned 4 to 6 percent after taxes which, although 
it may seem low, it is not out of line with earnings of CAB-certificated 
carriers in the same years.^^ And Southwest Airlines, now that heavy 
startup costs have been covered, is also profitable, having earned a re- 
turn on investment of 12.6 percent in 1974.^° 

How, then do the intrastate carriers achieve profits in combination 
with fares substantially lower than those charged by the CAB-certifi- 
€ated carriers ? The answer lies in a number of considerations, but most 
importantly in two : the seating capacity of the aircraft, and the load 
iactor. Thus, in the case of the Boeing 727-200 (one of the most widely- 
used aircraft ever built) PSA puts 158 seats in each aircraft, com- 
pared with 120-183 seats in trunk aircraft.^^ Does this mean that the 
passenger enjoys less space on intrastate aircraft? In the case of coach 
passengers, any space improvements provided by interstate carriers on 
short-haul routes are minimal, for the six-abreast seating with 34- 
inch ^2 pitch used by PSA is common among the trunk and local-serv- 
ice carriers on short hauls. The difference in seating capacity would 
seem to be accounted for by two things : first class service on trunk 
carriers (but not PSA), and galley space (but, as we have already 
noted, meal service to coach passengers on trunk routes is rather rare 
for the short hauls of relevance here) . Air California and Southwest 
Airlines (in Texas) get similarly larger numbers of seats into their 
Boeing 737's.«^ 

As regards load factors, all three intrastate carriers have consist- 
ently achieved higher load factors than have the interstate carriers. 
Thus, over the early 1970's, Air California achieved an average load 
factor of 70 percent ; PSA achieved a load factor of 60 percent or more 
on all its high-density routes, which in turn account for over two- 
thirds of its traffic. And, after its initial start-up period, Southwest 
Airlines achieved a load factor of 58 percent in 1974, and 62 percent 
in the first three quarters of 1975. The trunk carriers, on the other 
hand, have consistently achieved lower load factors of 52.1 percent in 

^ Coleman, p. T5. 

59 Ibid,, p. 75. 

«" Slmat, Hellliesen, and Eichner, vol. I, p. 8. 

«i Simat, Hellieson, and Eichner, vol. II, p. IV-4. 

«2 Pitch is the distance from a given spot on one seat to the same spot on the seat in 
front or behind it. 

«3 Air California puts seats into a 737-200 (Hearings of Subcommittee on Administra- 
tive Practices and Procedures, vol. I, p. 450), and Southwest puts 110 seats in the same 
aircraft (ibid., vol. II, p. 1243). On the other hand, United and Western's 737's seat only 
95. See Simat. Helliesen. and Eichner, p. 11-75. 


1972, 51.9 percent in 1973, 55.7 percent in 1974, and 54.8 percent in 

1975 ®* 

Itmight be argued that the lower load factor provided by CAB- 
reonlated carriers does produce a superior quality by making it easier 
to'^o-et a reservation at the preferred time.^^ If such benefits occurred, 
however, they accrued exclusively to the privileged ten per cent o± 
interstate passengers who flew first class. In 1974 and 1975 alike, the 
coach load factor was nearly 59 percent, roughly the same as that 
achieved bv PSA and Southwest.^^ 

Evidence from the intrastate markets, then, strongly supports the 
contention that interstate trunk fares are being set "artificially hi^h 
by the Civil Aeronautics Board, and that the potential excess prohts 
from these higli fares are being competed away through service qual- 
ity and flight frequency competition. 


The preceding analvsis makes an assumption which some might 
reo-ard to be una^cceptable— that less restrictive regulation would have 
the same effects throughout the United States as it has had on equiva- 
lent routes in California and Texas. It must be asked whether operat- 
ing and traffic conditions (i.e., weather, seasonality or traffic, conges- 
tion, etc.) on most interstate routes somehow make achievement of 
fares as low as those cited impossible. All these issues were investi- 
gated in a Report by the Subcommittee on Administrative Practices 
and Procedure of the Senate Judiciary Committee in 1975, and it 
concluded that none of these factors could account for a significant 
difference in fares. 

Nevertheless, it would be desirable to know whether the trunk car- 
riers could, if they adopted certain operating jn-ocedures of the intra- 
state airlines, such as seating density, achieve the same fares as the in- 
trastate carriers, operating in a less regulated environment. In earlier 
studies (Keeler, 1971, 1972), the present author has considered these 

The model was based on cost and operating characteristics of effi- 
cient aircraft with all coach configurations, indirect (non-flight) costs 
of low cost interstate trunk lines, and a 60 percent load factor. Essen- 
tially the same model is presented in section 3 of the appendix to this 
chapter. For hauls under 900 miles, seating configurations were as- 
sumed the same as for the intrastate carriers, with galley space for 
beverages and snacks, but not regular meals. For hauls over 900 miles, 
galley space adequate for full meal service was assumed.^® (The im- 
plications for our results of allowing full galley space on shorter 
hauls will be considered later) . 

** Intrastate load factors come from Simat. et al.. pp. 4, 8: interstate load factors come 
from U.S. Civil Aeronautics Board, Air Carrier Operating Statistics, December 1973 and 
December 1975. 

^'Thls matter Is discussed at length later in this section, under the discussion of Douglas 
and Miller's schedule delay model. 

* See U.S. Civil Aeronautics Board, Air Carrier Operating Statistics (December 1975) 
for interstate load factors ; PSA and Southwest load factors are mentioned above. A full 
discussion of the impact of CAB regulation on coach service quality is presented below 
in this section. 

^"^ Resource Allocation in Intercity Pa.^sengcr Transportation (1971). chapter 3, and 
"Airline Regulation and Market Performance." Bell .Tournal, Autumn 1972. 

^ For hauls over 900 miles, use of DC8-61 aircraft was assumed. Capacity was assumed 
to be 251 passengers, which does include galley spac« adequate for meal service. See 
Keeler (1972), pp. 406, 416. 


Because the California intrastate carriers had managed to achieve a 
load factor of over 60 percent up into the mid-1960's, it was assumed 
that loosened regulatory policies on the top of 30 routes in the United 
States would enable achievement of similar load factor elsewhere. 
Therefore, a 60 percent load factor was assumed for these calculations. 
To the extent that a higher load factor were achieved, either lower 
fares or higher carrier profits would be feasible. 

One test of the accuracy of the results of the procedure is its ability 
to predict California Intrastate fares, despite the fact that it is based 
on trunk costs. It, in fact, came quite close to predicting California 
fares. For the Los Angeles-San Francisco route, the estimated fare 
is $13.87 compared with an actual 1968 fare of $13.50. For the Los 
Angeles-San Diego route, the estimated fare is $6.50, compared with 
an actual fare of $6.35.^^ In each case, the model comes within 3 per- 
cent of predicting actual fares. For short hauls then, the model might 
give some indication as to what would be feasible elsewhere with more 
relaxed regulatory policies. Unfortunately, there are no unregulated 
long-haul markets with which to compare results. But, in April 1967, 
World Airways, a supplemental charter carrier, did attempt to enter 
the transcontinental market. In an application to the CAB, it pro- 
posed jet service between California and the East Coast with a fare 
of $79.50.'° This compared with estimated cost-based fares of $79.20 
on the Xew York-Los Angeles route, and $82.08 on the Xew York-San 
P^rancisco route. Thus, for three perhaps special cases, the model does 
a reasonably good job of predicting unregulated fares. 

On the unregulated routes, fares exceed predicted unregulated ones 
by margins ranging from 20 percent on the shortest hauls (such as 
New York-Boston) to over 90 percent in other cases, such as Xew 
York-Miami. Thus, CAB-regulated trunk fares exceeded estimates 
of feasible unregulated fares by 20 to 90 percent. 

In connection with the previously mentioned investigation of the 
Senate Subconmiittee on Administration Practices and Procedure, the 
results of this study were updated to the third quarter of 1974.'^ 
The results of this updating, again for the 30 top-density routes in 
the United States plus Los Angeles-San Diego, are shown in Table 4. 

For the 1974 updating, there are three intrastate markets on which 
to test the calibration of the model : Los Angeles-San Diego, Los An- 
geles-San Francisco, and Dallas-Houston. This time, actual intra- 
state fares are below predicted "unregulated'' fares by about 13 per- 
cent between Los Angeles and San Francisco and they are above hy- 
pothetical "unregulated" fares by less than 5 percent betAveen Los 
Angeles and San Diego. Between Dallas and Houston, the model would 
seem to be further off the mark; actual fares are above predicted "un- 
regulated" ones by 23.8 percent. But the fare used in the Table is 
the "peak" fare, valid between 7 a.m. and 7 p.m. In the case of this 
route, the ofl'peak fare of $13.88 is far less restricted than any other 
discount fare : it may be used after 7 p.m. on weekdays and anytime 
on weekends, and it is 25.7 percent below the estimated "unregulated" 
fare. A large fraction of passengers uses this latter fare, as well. 

«■' "Airline Regulation and Market Performance," p. 417. 
"'' JMd. 

-1 T Keeler. prepared statement. Hearings, U.S. Senate Subcommittee on Administrative 
Practices and Procedures, vol. II, pp. 1302-1305. muisLiauvt? 



City pair 









coach or 







jet discount 




(1) to (3) 


(1) to (4) 





Interstate routes: 

Atlanta to New York 44. 

Boston to Chicago 49. 

Boston to New York 17. 

Boston to Philadelphia 21. 

Boston to Washington 26. 

Buffalo to New York 22. 

Chicago to Cleveland 23. 

Chicago to Detroit 20. 

Chicago to Los Angeles 89. 

Chicago to Miami.. 64. 

Chicagj to Minneapolis 23. 

Chicago to New York 43. 

Chicago to Philadelphia 41. 

Chicago to St. Louis 19. 

Chicago to San Francisco 95. 

Chicago to Washington 37. 

Cleveland to New York 28. 

Detroit to New York 31. 

Las Vegas to Los Angeles 18. 

Lcs Angeles to New York 126. 

Los Angeles to Seattle 53. 

Miami to New York . . 62. 

New York to Pittsburgh 24. 

New York to Rochester 20. 

New Yjrk to San Francisco 133. 

Nevi/ York to Syracuse .. . 18. 

New York to Washington 19. 

San Francisco to Seattle 38. 

Intrastate routes: 

Dallas to Houston 18. 

Los Angeles to San Diego 9. 

Los Angeles to San Francisco 20. 




























































































































































Source: Fares come from Official Airline 1 (Oct. 1, 1974); California intrastate off-peak fares come from PSA flightsched- 
ule, Oct. 18, 1974; for derivation of other figures, see text. 

Overall, this model mi<>}it still bo oxpoctod to n^ivo roiiirli predictions 
as to fares in the absence of CAH reirtilatioii. An ninveiirhted average 
of the thiee intrastate markups Avonld indicate that the model has 
iinderpredicted inti'astate fares by no more than 6 ])ercent (iirnorino- 
off-peak fares in Texas for the time being). If data to calculate a 
weighted average were available, it wonld most certainly indicate a 
lower average markup yet, given that the Los Angeles-San Fi-ancisco 
ronte is far larger than any of the others. 

As regai'ds intei-state routes, it is interesting to note that the range 
of "markups" was not so broad by 1074 as it was in 1068: in 1074. it 
was oO to of) percent, rather than 20 to 00 percent. F'urthermore, in 1074 
the loAver markups were on the longer hauls — just the opposite of what 
was tnie in 1068. Overall, the situation had perhaps improved some- 
what on a number of longhaul routes, and tlie improvement is largely 
attributable to the results of the DPP'I, which included an effort to 
adjust the fare taper (i.e., decline in per-mile fare with distance) so it 
matched the taper of costs."^ But the evidence still indicates that fares 
were higher as of 1074 than they would have been without CAB 

An objection which can be made to all the comparisons made so far 
is that they are based on unrestricted, regular fares. In many cases, 

■"^ See Douglas and Miller, chapter 
83-944—78 8 


interstate travelers pay fares considerably below regular, unrestricted 
fares, because of discounts of various sorts. Thus, several studies have 
used realized revenues per passenger on a route as a measure of CAB 
fares, rather than actual fares, in comparison with intrastate fares/^ 

However, it is a mistake to compare revenue yields for CAB -regu- 
lated fares with regular fares for intrastate flights. The reason is sim- 
ple : discount fares generally impose travel restrictions : the trip must 
be made at inconvenient hours, with round-trip restrictions, or with 
some advance reservation restriction. Thus, a discount fare ticket is a 
different product from a regular fare ticket, as any businessman will 
testify. In evaluating the impact of regulation, it is obviously impor- 
tant to standardize for product quality, to the extent possible, and 
therefore, if regular fares are used for intrastate routes, they should 
be used for interstate routes as well. 

In any event, discount fares are available in intrastate routes, as 
well as interstate ones, and it is worth comparing the discount fares 
available on the two types of routes (it would be a mistake to compare 
revenue yields on the two route types, because product quality mixes 
differ) . This comparison gives further information on the relative fare 
levels in the two markets. 

Table 4 lists the lowest available discount fare on each route. The 
only criterion for inclusion is that the fare must not discriminate 
against any person on the basis of age or occupation, and it must be 
for a regularly scheduled flight with advance reservations accepted. 

One would expect off-peak discount fares to be lower than the unreg- 
ulated fares hypothesized here because the cost model assumes utiliza- 
tion rates consistent mainly with daytime-only operation. The marginal 
cost of off-peak operation should be considerably lower. The results in 
Table 4 indicate that this is true : in the three intrastate markets, the 
average discount fare is only 70 percent of the estimated peak or regu- 
lar fare (again, unweighted averages are used because of lack of intra- 
state passenger-mile data). On the other hand, in the interstate 
markets, the unweighted average discount fare is 20 percent higher 
than the estimated unregulated fare. This implies that, controlling for 
length of haul, the average discount or off-peak fare in interstate mar- 
kets was over 70 percent greater than an equivalent discount fare on 
intrastate routes (1.2 divided by .7 = 1.714). 

A recent study, done by the General Accounting Office to evaluate, 
extend, and update through 1974 the earlier study by the present writer 
(Keeler, 1972), confirms the results reported here."^* For the years 
1969-74, it finds existing fares which exceed efficient ones by similar 
margins, and its extends the analysis to all trunk airline routes in the 
United States. Overall, it finds that CAB regulatory policies over the 
1969-74 period cost travellers between $1.4 billion and $1.8 billion 
annually. And it is worth noting that all GAO fare comparisons are 
based on revenue yields, rather than published fares. 

Overall, then, it appears that CAB regulation, as of 1969-74, con- 
tinued to extract a toll in higher coach air fares than would have existed 
in its absence. 

" These studies include those of Douglas and Miller and Pustay, among others. 

''* romptroller General of the United States, Report to Congress: Lower Airline Costs 
Per Passenger Are Possible in the United States and Could Result in Lower Fares (Wash- 
ington, U.S. General Accounting Office, 1977). 



While the discussion so far has acknowledged that CAB regulation 
of interstate plane fares has some effect on product quality (flight 
frequency and load factor), it has taken no explicit, quantitative ac- 
count of these effects. The first work to analyze these dimensions of 
air transportation quantitatively was that of Douglas and Miller, pub- 
lished in 1974." This work attempts, in a most ingenious way, to quan- 
tify the benefits and costs of the quality changes which CAB regula- 
tion has brought about. The basic idea behind their methodology is 
that lower load factors allow more convenient service (in ways de- 
scribed below) , but cost the traveler more in fares. Their model trades 
off the value of this convenience to the traveler against the cost of 
providing it. 
The Douglas-Miller methodology 

They assume initially that every traveler has a "most preferred" time 
of departure. To the extent that he is unable to leave exactly when he 
would most prefer, he is subject to a "schedule delay" equal to the dif- 
ference between the time he would prefer to depart and the time he does 
depart. This schedule delay is made up of two components. First, it is 
unlikely that any one flight will be scheduled exactly at the desired 
departure time, even if the traveler can get reservations on that flight. 
The difference between desired departure time and the departure time 
of the nearest flight is termed frequency delay. Furthermore, the flight 
closest to the desired departure time may be booked. The average time 
incurred waiting because the nearest flight is book is termed stochastic 
delay. Schedule delay is simply the sum of frequency delay and 
stochastic delay. 

Douglas and Miller not only define the concept of schedule delay, 
but they also provide estimates of how much schedule delay is likely 
to occur on a given route per passenger (averaged over the year) as a 
function of the amount of traffic on the route, the size of aircraft used, 
and the number of flights (assuming a given seating configuration). 

Holding the number of passengers on a route and t]ic capacity of an 
aircraft constant, their method provides an estimate of expected sched- 
ule delay per passenger as a function of load factor. Now, a lower 
load factor costs more to provide in terms of operating costs, but it 
provides additional benefits in terms of reduced schedule delay time. 
The point of Doughas and Millers analysis is to estimate and weigh 
these benefits and costs quantitatively. 

To do so, it is necessary to evaluate in money terms both the costs 
of varying load factors and the benefits of varying schedule delay. 
The costs of offering various load factors and flight frequencies for a 
given trip, plan type, and seating configuration can most certainly 
be estimated, and Douglas and ^liller do so. However, it is much more 
difficult to evaluate schedule delay time, that is. to determine the 
amount which the average passenger would be willing to pay to leave 
a minute (or hour, or whatever) closer to his preferred departure time. 
Douglas and Miller assume that the typical 2:)assenger would value his 
schedule delay time at somewhere between $5 and $10 per hour, i.e., 

"" "Economic Regulation of Domestic Air Transport," chapters 5 and 6. 


they assume these as alternative values for their calculations. Tliese 
assumptions are probably somewhat on the high side, inasmuch as they 
are based on econometric estimates of the value of in-vehicle travel 
time, wherein the traveler is limited in the use he can make of that 
time.^^ Schedule delay time, on the other hand, can be used more or less 
pleasurably and/or productively because it does not have to be spent 
In a vehicle or even waiting at the airportJ^ For this reason, their re- 
sults based on a $o per hour value of schedule delay time are more likely 
to be valid than their estimates based on a value of $10."^ 

The output of the Douglas and Miller model consists of both optimal 
load factors and optimal fares, as a function of route, plane type, and 
seating configuration J^ The optimal fares are based on costs estimated 
in turn from a cross section of trunk airlines. 

Results of the Douglas-Miller mwdel — Optknal fare level 

On the basis of their calculations, Douglas and Miller provide an 
estimate of the welfare loss from CAB regulation, as of 1969. Their 
results are worth quoting in full : 

During 1969, . . . domestic trunk-line revenue totalled $6,514 million. Had the 
price-quality options been at optimal levels consistent with a $10 per hour eval- 
uation of passenger delay time, total passenger revenue would have been approx- 
imately $6,148 million ; for an optimal configuration reflecting a $5 per hour 
passenger evaluation of delay time, revenue would have been approximately 
$5,976 million. In other words, during 1969 passengers paid excess fares, rang- 
ing from approximately $366 million to $538 million. 

However, for this additional price, passengers purchased reductions in delay 
time — valued at $118 million in the $5 per hour configuration consistent with a 
$10 per hour waiting time evaluation, and approximately $182 million in the .'*?5 
per hour configuration This leaves a 1969 deadweight welfare loss which (given 
the assumptions) ranges between $248 million and $356 million.**^ 

After quoting this, it is necessary to point out also that Douglas and 
[Miller qualify their welfare loss results by asserting that they are ^'just 
not that important" relative to the basic notion they introduce that 
service quality considerations are important to calculations of the wel- 
fare impact of airline regulation. The point remains that service qual- 
ity can be "too high" — that it reduces overall schedule delay below 
levels for which customers, given the choice, would be willing to pay. 

Furthermore, Douglas and Miller's numerical results are of consid- 
erable interest, because they imply a welfare loss from CAB regula- 

•8 The demand studies from which Douglas and Miller get their time values are Arthur 
De Vany. "The Revealed Value of Time in Air Travel." (Review of Economics and 
Statistics 56 (February 1974), pp. 77-82; and S. Brown and W. Watkins, "Measuring 
Elasticities of Air Travel from New Cross-Sectional Data," delivered at annual meetings 
of American Statistical Association, 1971. 

''^ If a pa>^senger makes reservations, he knows when his plane will leave, and need 
only go to the airport just before it does. Thus, in the case of airline schedule delay, 
"waiting" simply amounts to changing one's plans so as to do something other than 
traveling at the most desired departure time. 

'8 Furthermore, if plane fares are currently higher than they optimallv would be. it 
is more or less certain that if fares were lowered, new travel by those w'ith lower time 
values would be induced to travel, drawing the average value of schedule delav time 
down further. As a result, it will achieve inappropriate results to use time values esti- 
mated only for those currently traveling by plane. 

™ As route density rises, optimal load factor rises also, because as route densitv goes 
up. the extra benefits of frequency do not rise proportionally to tlie number of passengers. 
A longer haul tends to reduce the optimal frequency. For the reasons behind this, see 
the subsection on optimal fare structure below. 

80 "Economic Regulation of Domestic Air Transport," p. 72. The welfare loss which 
they calculate is lower with a higher value of time because the low load factors which, 
the CAB has brought about are most convenient for travelers with high time values — 
those willing to pay a high fare with the corresponding reward of easier reservations 
and less schedule delay. 


tion of only 4 to 6 percent of total fares charged,®^ whereas prev-ioiis 
studies of the welfare costs of CAB regulation for the late 1960's and 
early 1970's would put the number at anywhere from 20 to 55 percent 
of interstate fares charged.®- 

To some degree, this difference stems from the fact that Douglas 
and Miller have subtracted the benefits of improved service quality 
made possible from CAB regulation from the costs in excess fares. 
However, that does not explain the whole difference, for their results 
imply that optimal fares are no more than 6 to 9 percent lower than 
actual ones on CAB-regulated routes. 

The main difference between Doughis and Miller's welfare loss esti- 
mates and those im])lied by previous studies would seem to stem from 
dilTerences in measurement of costs and fares. Let us consider measure- 
ment of fares first. 

To measure actual fares. Douglas and Miller use not official fares, 
but instead revenue yields, taking account of discounts and the like 
to the extent that tliey aie used. This has a problem, however, for the 
restrictions attaclied to discount fares make the trips attached to them 
inferior products, as discussed above. Therefore, it is at least as reason- 
able to compare actual regular fares on a given route with optimal 
ones, assuming that the 'V)])timal'* fares would have no strings attached 
(and indeed, the much lower regular fares available in intrastate 
markets do have no stiings attached, except for the rock-bottom night 
and weekend fares in Texas). 

'Slave importantly, it appears that Douglas and Miller have over- 
estimated airline costs, at least relative to what they are on the Cali- 
fornia and Texas routes. For example, their cost estimates project that 
in 1971. the year for which their costs are calibiated. PSA would lose 
over $2.50 per passenger at a 100 percent load factoi-. and over $0 per 
passengeT', or 40 percent of the fare chaiged. at a more realistic 70 
percent load factoi-.^"^ Details of these comparisons, plus a more de- 
tailed critique of Douglas and Miller's cost estimates, are contained 
in section IT of the appendix to this study. 

DaugJa'9 anrl M'tllcr theory : Optimal fare strvctiire 

Douglas and Miller's model analyzes not only the optimal fare level, 
but also the optimal structure of faies. i.e.. the way they should vary 
by length and level of traffic for a given route. Tliis requires further 

As pi-eviously stated, thoir model trades off the benefits of additional 
flight fiequency on a iriven route (with a iriven flow of passengers per 
year) against the costs of that additional flight frequencv.^* But the 
benefits and costs of flight fiequency are not independent of trip 
length and route density (i.e.. flow of passengers per unit of time 

^^Tliat- is. total rpvoniios In lOfin wpfp Sfi.l4« million, and Donslas and ^flller'^ pstl- 
mate? of the costs of regulation were .<243-.Sr)6 million, or 4-o percent of the total. 

82 For example, tut own estimates of the welfare costs of CAB recnlation on the sample 
studies of SO routes In lf)6S was 20-f)0 percent of the efficient fares. This ig equivalent 
to savlncr that the welfare loss is 17 — i7 percent of eiistlntr fares (1/1.2=.83. and 1— 83= 
.17: similarly. l/1.9=.r.3. and 1 — .53=.47i. 

» These calculations are based on Tables 2-7. p. 24 of Douglas and Miller. Further 
details of these calculations are provided In Section II of the appendix to the chapter 
of the present work. 

^ This anal.vsis Is done holdlnsr number of passengers on a route constant. In general. 
a higher fare does not necessarily result In a higher frequency, despite the fallacious 
assertions of many that that is the case. The effect of a change In fare on flight frequencv 
depends on the change on the number of passengers as a result of the fare change. 


traveling on the route). More specifically, as the length of an airline 
flight increases, the costs of providing additional flights increase as 
well. The benefits of this additional frequency, however, do not in- 
crease : though in-vehicle travel time increases with length of flight, 
schedule delay time is a function of frequency, load factor, and plane 
size alone. Therefore, as the length of haul (and hence the cost of a 
flight) increases, it will generally be most efficient to trade off some 
of the higher cost of flights for more schedule delay time. In other 
words, as flights get longer in haul, it will pay to reduce flight fre- 
quency because long-haul flights are costlier than short-haul ones, on 
a per-trip basis. Therefore, optimal fares will entail higher load factors 
on longer hauls. Before the DPFI, CAB fares did just the opposite : 
the breakeven load factor declined as loncrth of haul rose.^^ 

Another element of Douglas and Miller's theory of optimal fare^ 
structure relates to route density. As route density (again, the number 
of passengers traveling on a route per unit of time) increases, one- 
would expect flight frequency to rise, as well. But Douglas and Miller 
show that it is not efficient to increase flight frequency proportionally 
with the number of passengers traveling on a route. This is because 
of the indivisibility of aircraft, i.e., on low density routes, aircraft 
cannot be made smaller than a certain size (say, 110 coach seats) with- 
out a sharp increase in operating costs. Maintaining an optimal fre- 
quency on a low-density route may thus require operating 110 seat 
planes at low load factors. As route density rises, the optimal strategy 
is to fill up more seats, so that optimal flight frequency does not rise 
proportionately with route density. Given that the benefits of flight 
frequency do not increase proportionally with passengers, the optimal 
load factor should be expected to rise with route density. 

Currently, all CAB fares are based on a 55 percent load factor, 
standard, independent of route density and of length of haul.^^ Be- 
cause of this, CAB regulation imposes inefficiencies from fare struc- 
ture in addition to those stemming from fare level. 

Douglas and Miller's welfare loss estimate includes the costs of 
nonoptimal fare structure with respect to length of trip, but not with 
respect to route density. In subsequent work, however, based on Doug- 
las and Miller's methodology, M. Pustay has attempted to estimate the 
cost of nonoptimal fare structure, as well as nonoptimal fare level.^' 
His estimates will be considered later in this section. 

Overall, Douglas and Miller's methodology of estimating optimal 
fares and service qualities is a most important contribution. But we 
shall attempt here to get estimates of the welfare loss from CAB regu- 
lation incorporating their model of schedule delay, but based on a 
cost model calibrated to fit more accurately the costs and fares achieved 
in the less-regulated environments of California and Texas, and to get 
at least a rough idea as to what additional costs are imposed by non- 

85 This is equivalent to saying that the ratio of fare to cost at a constant load factor 
rose with distance. That this was so may be seen from Keeler, "Airline Regulation and 
Market Performance," Table 8. 

8« Alfred E. Kahn, "The Regulation of Air Transportation — Getting From Here to There'^ 
in Regulating Business- The Search for an Optimum, Institute for Contemporary Studies, 

87 "The Effects of Regulation on Resource Allocation in the Domestic Trunk Airline 
Industry," chapter 3. 


optimal structure in CAB fares not accounted for in Douglas and 
Miller's published work. 

Before doing that, however, it is important to consider another issue. 
The argument for deregulation hinges implicity on the assumption 
that competition will generate an optimal combination of fare and 
service quality. It is now time to examine the evidence as to whether 
this is so. 


As previously stated, competitive markets can fail to achieve 
economically efficient results because of three broad types of problems : 
monopoly power at the firm level caused by increasing returns to scale, 
externalities at the market level, and imperfect information about the 
market on the part of consumers and/or producers. It is worth con- 
sidering the likeliliood of each of these i^roblems occurring in the trunk 
airline industry. 

Returns to Scale for Ti'unh Airlines 

Many statistical studies have been done b}' economists investigating 
the extent of increasing returns in the airline industry. '^^ The metliodol- 
ogies differ considerably from study to study, but the basic question 
asked is the same in each case : what happens to airline operating costs 
per miit of output as firm output rises (and as firm size adjusts these 
changes in output) 'i 

The most common way to answer this question is via statistical 
analysis of a cross section of firms, measuring how costs vary as one 
goes from small firms to large firms. In the simplest such studies, 
output is measured with only one variable, and it is possible to 
determine scale economies visually by way of a ''scatter diagram,'' i.e., 
a diagram plotting firm costs per unit of output (average costs) 
against output. If average costs are constant with output, then there 
are constant returns to scale, and there is no reason to expect market 
failure, for reasons discussed above. Similarly, if average costs in- 
crease with firm output, that is similarly consistent with efficient, com- 
petitive markets.^^ However, if average costs decline with firm output, 
that is a sign of increasing returns to scale, and under those circum- 
stances one should expect market failure, for reasons discussed in 
section III. 

One recent study of returns to scale in the airline industr}', by the 
CAB Special Task Force on Eegulatory lieform, has measured re- 
turns to scale in a two-dimensional, graphic way, and the evidence it 
finds is worth showing here as wcll.^" 

^ Among the studies in footnote 1 which considered returns to scale in the airline 
industry are Caves (chapter 3), Keeler (indirect costs only. pp. 406-411), and Douglas 
and Miller, chapter 2. Also considering scale economies implicitly (by comparing effi- 
ciency measured in various ways for large firms and small firms) are Jordan, mentioned 
in footnote 1, and Robert J. Gordon, "Airline Costs and Managerial Efficiency," in J. R. 
Meyer, Editor, Transportation Economics: A Conference (New York. National Bureau of 
Economic Research, 1965), pp. 61-94. Finally, Pulsifer, et al., measure scale economies, 
as discusijcd below in the text. 

^8 It will be recalled that destructive competition and natural monopoly occur only 
"When marginal cost is below average costs. If average cost is constant or increasing, then 
the cost of the last unit of output will always be equal to or greater than the average 
cost of all units, so marginal costs will be equal to or above average costs. 

^ Pulsifer, et al., pp. 102-107. 




Locals = X 
Trunks = • 

12 3 4 5 6 7 8 

Ayilable Ton Mi/es (Billions) 

Figure 1 
Source : Pulsifer, et al., p. 104. 


60 - 
55 - 



Locals = X 
Trunks = • 




300 400 

Routt Dtnsity 
ATM Per Weighted Average Route Miles (ThouioryJs) 

Figure 2 


Source : Pulsifer, et al., p. 105. 



What the Task Force did is to graph each of two different measures 
of airline output against cost per unit of output, based on data for the 
trunk and local service carriers for the fiscal year (ending June 30) of 
1974. The two alternative firm output measures used are total avail- 
able ton-miles per firm per year (an available ton-mile being the total 
tons of aircraft capacity transported times the distance each ton of 
capacity was carried, summed over all the trips), and available ton- 
miles per route-mile. Scale economies defined with respect to both out- 
put measures, are relevant here : use of the available ton-miles meas- 
ures economies of firm size (i.e., whether large firms are more efficient 
than small ones). Use of route density measures economies of high- 
density operation versus low-density. If there were economies of route 
density, then the marginal cost of adding traffic to a firm on a given 
route would be less than average cost, and there would clearly be 
potential for short-run restructive competition, leading to monopoly. 

The resulting scatter diagrams found by the CAB Task Force are 
shown in Figures 1 and 2 which show, respectively, economies of firm 
size and economies of traffic density. 

As regards firm size. Figure 1 indicates tliat once the size of the 
smallest trunk line is reached, average costs are constant and inde- 
pendent of firm output. To put it another way, the average cost cun^e, 
which giaphs average cost and output (as fitted to the scatter of 
points) is fiat over the entire range of output for trunk carriers. 

At first glance. Figure 2 would seem to indicate economies of route 
density at the output levels of the smallest trunks, based on the line 
drawn through the scatter of points. Xote, however, that the rise in 
that curve is due exclusively to the higher costs of the local service 
carriers (repTCsented on the graph by x's) at the lowest route den- 
sities: a line drawn to sninniarize the dots which repi'eseut the trunk 
lines would be perfectly flat over the full range of traffic densities. The 
higher observed costs of the local service carriers in this graph would 
appear to be due to the fact that they operate at a shorter average 
length of haul than do the trunks, and the graph does not control for 
that fact. 

In short, these gi-aphs give stroiig evidence of constant returns to 
scale for domestic trunk airlines airlines, over a very wide range of 
outputs. The graphs shown, however, have the severe disadvantage 
that they cannot show more than two variables fcost and output) at 
the same time. To control for other variables (such as length of haul), 
and to allow for more than one dimension of output passengers and 
other traffic variables, as well as ca])acity varial)les such as available 
ton-miles), it is necessarv^ to estimate more elaborate multi-dimen- 
sional statistical equations. More elaborate studies have been done, 
however, and they are unanimous in supporting the result of constant 
returns to scale, however scale is measured.®^ 

However, there is another type of scale economy in scheduled trans- 
portation which could cause market failure in the airline industry. It 

" These include all the studies mentioned in footnote 85. One other studv, for local 
service airlines, did find scale economies. See George Eads. Marc Nerlove, and William 
Raduchel. "A Long-run Cost Function for the Local Service Airline Industrv." Review 
of Economics and Statistics 51 (August 1969). pp. 258-270. It would thus appear that the 
minimum eflBcient scale (the smallest size consistent with minimum-cost operation) in- 
airlines lies somewhere between the largest local service carrier and the smallest trunk 


stems not from scale economies in airline operations, but rather from 
what happens to the quality of service faced by passengers as route 
density increases. The argument goes as follows : as passenger traffic 
density increases, so eventually must frequency of flights, for a given 
aircraft size. With more frequent flights, people will be able to leave 
closer to their exact preferred departure times, and schedule delay 
costs will be reduced.^^ (It is worth noting that these economies accrue 
to the passengers directly, rather than to the firms.) 

In that sense, social costs of scheduled transportation decline with 
output, marginal costs are below average costs, and marginal cost 
pricing will require a subsidy.^^ 

On the densest routes (say, the top 100) there is evidence that an 
extra flight (holding load factor constant) reduces total trip time by a 
trivial amount, and this problem is unimportant.^^ As for the less- 
dense routes, there is evidence that the optimal subsidy may not be 
trivial — it could be ten per cent, fifteen per cent, or more of the optimal 

Still, it is not clear that CAB regulation has brought these less- 
than-top-density routes closer to an economically-efficient equilibrium 
point than would occur with complete deregulation. To the contrary, 
evidence from lower-density routes (relative to interstate trunk 
routes) in California and Texas would seem to indicate that these 
routes receive greater flight frequency at lower fares than similar 
interstate routes.^^ Furthermore, the prospect of a policy which sub- 
sidizes trunk air service to achieve an optimal equilibrium is not an 
appealing one — air transport is generally consumed by the relatively 
well-to-do, so subsidies would be highly regressive.^^ Of course, if 
subsidies are necessary to preserve service to low-density routes, the 
local service carriers exist to provide it. No currently-proposed regu- 
latory reform suggests elimination of these subsidies.®^ 

There is yet another reason why one might be skeptical of the notion 
of subsidizing low-density routes because of service-related economies 
of traffic density. It can be shown that the amount of the optimal sub- 
sidy depends crucially on the valuce of schedule delay time assumed 
for the typical air traveler.^^ The latter value is very difficult to meas- 
ure, and while one might argue that it should be used to determine 
the optimal tradeoff between flight frequency and fare, it may not be 
appropriate to base payment of taxpayers' money on a value measured 
with so little certainty. 

^- The pxnct formnla for Douglas and Miller's schedule delay function may be found in 
equation (4) of the appendix to this chapter. 

f*^ Snenifically, the amount of the subsidy for each passenser traveling should he the 
difference between marginal cost and average cost. This is because the optimal fare is 
marginal cost, but the firms must earn average cost to break even. With scale economies, 
marclnal cost is below average cost, it will be recalled. 

p-* See Hary .T. Dorman. Airline Competition r A Theoretical and Emnirical Analysis, 
(Ph. T). dissertation. University of California, Berkeley, 1976), chapter 7. 

^■> Tbid. 

^ Kennedy Renort, pp. 46—4,9. 

^"^ This is. in fact, an excellent reason for not subsidizing the local service carriers, but 
that sreneral topic is beyond the scope of this study. 

f's In fact, the Cannon-Kennedy bill, recently passed by the U.S. Senate, guarantees 
of government-subsidized service to any communities which would otherwise lose 
service after regulatory reform. See the final section of this report below. Such subsidies 
would be regressive, but if they are necessary to achieve regulatory reform because of 
political considerations, they are probably worth having, 

«• The reason for this is that the amount of the optimal subsidy depends on the rate at 
which costs (including the value of time saved) decline with traffic density : but this 
depends on the value of time, for it is only with a value time that one can estimate the 
cost reduction from additional frequency. 


Overall, then there is good reason to prefer less-regulated, but still 
unsubsidized air service on all trunk routes to what is available now 
with CAB regulation, even if one might hope to do a bit better yet by 
providing even lower fares than would exist with deregulation on low- 
density routes, supported with government subsidies. 

The potential for destructive competition 

There is another reason why decreasing costs (i.e., increasing returns 
to scale) have been cited as causing need for regulation: the possibil- 
ity of destructive competition. With decreasing costs, marginal costs 
are below average costs, and if non-collusive pricing drives price down 
to 7Tiarginal cost, all firms will lose money. 

To what extent are the service-related economics of traffic density 
just discussed likely to cause destructive competition? The answer to 
this question hinges on the extent to which these economies of high- 
density operation are realizable to the firm (or internal to the firm, 
in tlie jargon of microeconomics). In other words, will a firm with 
a large part of the market be able to charge a lower fare (or offer 
better service) than a firm with a small market share? Clearly, if a 
firm with a large market share can offer a lower total trip cost than 
ti firm with a small market share, something akin to destructive com- 
petition will occur : firms will cut fares and expand service in hopes of 
winninir over the entire market, and if one firm won a larger share 
than others, it would be in a position to drive the others out (because 
it would be able to offer a lower total trip cost). If this were the case, 
airline service would be a natural monopoly on any route where 
sei'vice-related economies of traffic density occurred. 

To what extent, then are these service-related economies of traffic 
density internal to the firm? It is possible to tell two stories to answer 
this question, and each story drives a different answei*. We shall con- 
sider each story before attempting to answer the question. 

Consider first the storv sugorogting that these economies are indeed 
internal to the firm.^"° This view assinnes that if fares across airlines 
are equal, a passenger selects an airline before he selects a flight; that 
is, he determines which airline ofTers the greatest frequency, assumes 
that that airline is likely to offer him the lowest schedule delay, and 
calls it, without even considerinir frequency levels of other airlines. 
In this case, if several airlines initially attempt to compete, the air- 
line whir'h offers the grea<"est frequency would have advantage relative 
to all others, for it would attract a disproportionate amount of the 
traffic (i.e.. a larger percentage of passengers than of fliirhts), it would 
have hiirher load factors, and in a world of deregulated prices, it 
would be able to undercut airlines with less frequent services in price 
and drive them out of business, as dosci'ibod above. ^^^ 

The first storv, however, is not based on totally rational consumer 
behavior. Especially in a lower-density route, where service-related 
economies of traffic density exist, a rational consumer will examine (or 
ask his travel agent to examine) all possible departures in the Official 

T^ A much mnro dotnilpd morlpl. baspd on this story. Is presontod in Dorn->nn. Hiapfpr .^. 

^"^^Thjs oonlfl nlso occur In a prlco-rcaulatcd market: the airline with the rrreatest 
freoiien.'^v wonld bo nhie to offer the lowest schedule delav time, nnd this wonld nid i*- 
in winninc more p.nssencrers. which would irive it an even lower schedule delnv time, and 
CO on This arjriiment is develooed theoretically in .Joseph V. Vance. "The Possihilitv of 
Loss-Prodncinjr Eonilihria in Air Carrier Markets," unpublished, 1971 ; Vance's model is 
summarized by Douglas and Miller, pn. 49, 59. 


Airline Guide before deciding which airline to use for a given leg of 
his trip. If this is done, the service- related economies of traffic density 
become external to the firm, because no airline will be able to attract 
a disproportionate amount of the traffic by offering a greater flight 
frequency than everyone else. 

Because it is based on a more rational traveler behavior than the first 
story, the second story, implying that service-related economies of 
traffic-density (when they exist) are external to the firm, if prefer- 
able to the first on a priori grounds. 

But we need not rely only on a priori assumptions of consumer ra- 
tionality to support the second story. There is also some empirical evi- 
dence. More specificall}^, note that if the first story were correct, firms 
with greater flight frequencies would have a larger share of passengers 
than of flights, because a disproportionate fraction of passengers 
would prefer the airline with the greatest flight frequency. This in 
turn implies that, with constant fares across firms, firms with more 
frequent flights would have higher load factors (and higher profits, 
given constant fares) than firms with less frequent flights. ISTote, fur- 
thermore, that all these arguments hold with or without fare regula- 
tion: they are in no way dependent on deregulation. Thus, a study 
which compared market shares (in terms of flights) and load factors 
in various markets (including those regulated by the CiAB) should 
shed light on this issue. 

To put things another way, if a statistical study across routes (or 
over time) found that firms with high market shares had higher load 
factors than firms with low market shares, that would be evidence of 
potential for destructive competition leading to monopoly. Yance has, 
in fact, done a statistical study of the relationship between market 
share and load factor for U.S. trunk carriers on a number of routes, 
and he has concluded that over a wide range of route densities, load 
factors are independent of market share. ^^^ Miller, in more recent 
work, has found additional and stronger evidence to the same effect.^°^ 
This supports the second model, which in turn implies that service- 
related economies of traffic density are external to the firm. 

On the lowest-density routes, service consideration could still create 
a natural monopoly: if potential traffic is so low that each airline's 
plane ends up leaving on a different day, that may create sufficient 
inconvience to make the traveler prefer one airline. 

But even in the case where there appears to be a "natural monopoly" 
on a low-density route, it is not clear that regulation is needed to 
achieve good market performance (i.e., price at or near marginal 
cost) : low-density routes tend to be short-haul in nature, so surface 
competition is strong. Furthermore, there is always the threat of 
entry of new firms if the "monopolist'' carrier sets his fares at too 
high a level, or provides service quality too low (this point is dis- 
cussed further in the final section of this study) . 

Exfenialities of Varimis Types 

Another potential cause of market failure in classical microeconom- 
ics stems from externalities of other sorts. For example, aircraft pol- 

ios ibid. 

103 James C. Miller III, "Airline Mnrket Shares and the Possibility of Shortnm I^ss 
Equilibria," presented to meetings of the Western Economic Association), June 10, 1974. 


lute the air and make noise, and the market in and of itself takes no ac- 
count of this. As a result, if aircraft noise and pollution go untaxed or 
uncontrolled, too much air service will be produced for economic effi- 
ciency. However, as is well-known, these environmental aspects of 
air transportation are reflated, and no one proposing deregulation 
is referring to these aspects of regulation, which are, in any event, not 
practiced by the CAB. 

Another sort of externality is what one might call '"location" exter- 
nalities comiected with the provision of scheduled transportation. It is 
akin to the location problem of. say drugstores : since some houses are 
closer to some drugstores than to others, and since traveling to shop 
has a cost, some drugstores will have a market advantage over others, 
at least for some customers. It has been shown some time ago that 
this will result in too many drugstores, operating with excess capac- 
ity at "too high*' a cost.^^'^ 

The situation with airlines is totally analogous, except that the 
competition is in time rather than space: airlines will all tend to 
cluster their departures around the same times (usually the most popu- 
lar ones) rather than spreading them out as would be more socially 
optimal. The theoiy behind this for airlines and other transporta- 
tion companies has been developed by Douglas (197-2). ^'^^ However, 
Douglas's argument depends on the assumj)tion of scale economics of 
plane size, i.e., lower operating costs (or a given length of haul) as 
the plane size gets larger. Once one srets past the capacity of a 727-200 
(150-160 passengers) such economies would appear to exist more on 
engineers' drawing boards than in reality. Thus, for a civen length 
of haul, Douglas and Miller ^et a higher cost per seat-mile for a 
Douglas DC-IO than for a Boeing 727-200.^06^ ^ 

It is probably true that this soii" of competition will cause a modest 
amount of market inefficiency here, especiallv on low-density routes. 
But to claim that regulation is needed to solve this problem is equi- 
valent to claiming that retail trade should be reirulated l>ecause of the 
similar problem there. ^"^ Most economists and most of society as a 
whole would seem to believe that the cost of this market failure in 
retail trade is too small to merit regulation there. Analogously, it is 
impossible to find any evidence that market failure here is great 
enouirh to merit regulation on this count in the case of airlines. Indeed. 
CAB regulation does nothino: to solve this problem, because it relates 
to plane size and the spacing of flifrhts over the time of day — some- 
thing the CAB has scrupulously avoided regulating.^''^ 

if^ "Too hipli a cost*' herp oconrs bpcansp of a firm sizp too small to miniml7;p rosts. See 
Edward rhambprlln. The Theory of Monopolistic Competition (Cambridge, Harvard Uni- 
versity Prpss. 19o0). 

ifK GporcTP Donprlas. "Priop 'Rperulation and Optimal Sprvicp Standards." Jotn-val of 
Transport Economics and Policy 20 (May 1072). pp. llf>-127. Also "Equilibrium in a 
DprPCTilntPd Air Transport Model." prpsented at Seminar on Problems of Regulation and 
Public T'tilitles. Dartmouth College, August 1972. 

^"^ See r>one:las and Mlllpr. p. 24. 

^'^ In rptnil trade, rpsmlation to tnkp account of this effect would involve forcing stores 
to spread out. If anything, the e.xisting location regulation for retail trade dops the oppo- 
site — it encourages clustering of trade activities through zoning. Presumably the un- 
pleasant "neighborhood effects" of having commercial activity on every block (or spread 
out in some such way) outweighs the effect discussed in the text as regards regulatory 

1* Basically, to control for this problem, a regulatory agency would attempt to coordi- 
nate scheduled departures of the carriers, spreading them out, and it would encourage 
the use of aircraft on a given route somewhat larger than the market would generate. 


There is another sort of market failure which might occur in the 
airline industry. The theory, of competitive markets assumes that 
consumei-s have perfect information about the quality of the product 
they are buying, or that they can learn by trial and error. But in 
the case of an airline, "learning" whether it is safe can be so ex- 
pensive as to cost the passenger his life. Furthermore, there is an 
externality: if one airline has unsafe procedures or aircraft, it can 
harm passengers or other airlines by midair collisions and the like; 
and crashes can, of course, cause damage and injury to those not 
flying. Thus, there is ample reason to favor continued safety regula- 
tion by the Federal A\dation Administration. But this has nothing to 
do with fare regulation by the CAB, and indeed, the evidence pro- 
vided by the intrastate carriers would indicate that they haxe excellent 
safety records, even in relation to CAB-certificated carriers.^^^ 

Further evidence on the safety of non-CAB-certificated carriers 
from the U.S. supplemental (i.e., charter) carriers. On the basis of 
fatalities per billion passenger-miles, the supplementals have, in fact, 
been safer than the trunk and local service carriers for nearly every 
year of their existence except for 1966 and 1970.^^° Overall, then, there 
is no evidence that deregulation of fares and entry is likely to lead to 
less safe operation, given continued safety regulation by the Federal 
Aviation Administration. 

Entr-y Barriers 

As is evident from the discussion so far, entry, or the threat of it, 
by new firms is extremely important to the functioning of competitive 
markets. Without it, there is no assurance that prices will be driven 
down to costs whenever excess profits are earned. If for some reason it 
is difficult for new firms to enter an industry, this may cause market 
failure, i.e., prices well above costs, and excess profits on the part of 
firms in the industry. 

In determining the viability of competitive markets in the airline 
industry, it is therefore important to consider what barriers to entry 
by new firms may exist, assuming that the current barrier of CAB 
regulation is lifted. Two types of entry are relevant here : entry^ by 
an existing scheduled airline firm into a new market (i.e., a new city- 
pair route) and entry into a given market by a firm not previously 
providing air transportation. We consider the case of a totally new 
firm first. 

Entry hy New Firms into the Airline Industry 
Following the earlier work of Bain and Caves,^^^ we consider three 
possible entry barriers to new fixrms in the airline industry : minimum 
capital requirements for efficient operations, absolute cost disadvan- 
tages to new firms, and product differentiation. 

Minimum capital requirements in an industry may occur because 
scale economies require a threshold investment to achieve efficient op- 
erations. If this investment level is high enough, it can make it dif- 
ficult for new firms to raise the capital needed to begin efficient 

109 See footnote 55. 
no Wennedy Report, p. 73, 

"iJoe S. Bain, Barriers to New 'Competition (Cambridge, Harvard University Press, 
1956) ; also Caves, op. cit., chapter 4, 


In the case of the airline industry, we haA'e already shown that scale 
economies are likely to be unimportant once a firm is as large as a 
trunk carrier. Still, Figures 1 and 2 indicate that for firms smaller 
than the trunk carriers, costs may be higher. The matter which con- 
cerns us here is just how large an investment is needed to achieve the 
minimum efficient airline size. The intrastate carriers in California 
and Texas provide some evidence here : we have already presented evi- 
dence that they have managed to charge fares considerably lower than 
the trunks on equivalent routes, and that their costs (controlling for 
such things as length of haul and load factor) are no higher than 
those of the trunks, and they may be lower. 

Consider first the case of PlSA. In the early 1960's, it managed to 
undercut trunk fares and earn a return on equity in excess of 20 per- 
cent (1960-64), using a fieet of only four to six Lockheed Electra 
aircraft. ^^^ 

More recently. Southwest Airlines instituted similar low-cost serv- 
ice in Texas, also undercutting trunk fares as previously described,, 
starting with only three Boeing 737-200 aircraft. In 1974, Southwest s 
total net physical assets under $16 million in value.^^^ This in- 
vestment level represents an extremely low barrier to entry relative 
to many unregulated industries in manufacturing, such as steel, where 
efficient, integrated plants cose in the biUions of dollars.^^* Capital 
requirements for new firms in the airline industry, then, are very low 
relative to those of many manufacturing industries which society has 
decided are sufficiently competitive not to merit regulation. 

The second possible disadxantage to new firms in tiie airline indus- 
try is an absolute cost disadvantage for a new firm: the idea is that 
existing firms may have ''learned ' to produce their services at lower 
costs than newcomers could achieve, the residt being that newcomers 
are at a cost disadvantage. If this were the case in airlines, it would 
obviously discourage entry. But again, the evidence from the intra- 
state carriers gives no indication of any such cost disadvantages. It is 
true that for their first two years or so of operation, both Air Cali- 
fornia and Southwest Airlines did lose money, as they got their serv- 
ices established.^^^ But initial losses were quite transitory, and, as 
previously stated, they were quickly supplanted by profit levels for 
both carriers significantly higher than those for the trunk carriers.^^^ 

The third possible disadvantage to a new firm in the airline industry 
steins from product dili'ereiitiation. That is, if existing firms can 
create the impression in the minds of the traveling j^ublic that their 
services are superior to those of potential new firms, this will create 
an entry barrier to those potential newcomers. Most certainly, airlines 
attempt to di lie rent late their products from those of other carriers, 
through the luxury of their service, the quality of their meals, the at- 
tractiveness of their stewardesses, or other considerations. The question 
is not, however, whether airlines attempt to differentiate their prod- 
ucts, but rather how ditferentiable the product inherently is. To put 
the question another way, how easy is it for one airline to copy an- 

11^ Jordan, p. 193. 

"3 Simat, Helliesen, and Eichner, vol. II. Appendix S. p. 2. 

"* See Bain, op. cit. and Leonard \V. Weiss, Case Studies In American Industry (New 
York: John Wiley, 1971), pp. 147-151, 

"5 See Simat, Helliesen, and Eichner, vol. II, part II and III. 
"8 See the lirst part of this section. 


other's selling innovation ? A priori^ it is difficult to see anything (serv- 
ice-wise) which one airline can do that another one cannot. Among 
different airlines, the aircraft are the same, the safety standards are 
the same, the flight crew training standards are the same, and tho 
ground service (i.e., reservations and baggage) technologies are the 
same. On the basis of available evidence, then, we have every reason 
to concur with the conclusion of Caves regarding the airline industry: 
"we find almost no characteristics in the product to lead us to expect 
that it would prove readily differentiable." ^^^ 

Entry hy Existing Airlines on to Neio Routes 

On the basis of the evidence presented above, the airline industry 
would be one of very low entry barriers relative to many manufactur- 
ing industries, even if all entry had to be by firms totally new to avia- 
tion. In fact, however, that is not the only way in which new entry can 
occur. It is possible that airlines already possessing scheduled routes 
can enter new routes, and it is possible for carriers now providing only 
charter or freight service to enter into scheduled passenger service. As 
a result of this, entry barriers into any one airline market are even 
lower than the immecliately-preceding discussion would imply. 

More specifically, the costs for a carrier to add a route to its net- 
work are low indeed : if a new city is served, it need only rent space in 
the airport, and hire appropriate ground support personnel. It might 
also initially advertise its new schedules intensively. 

But if the airline is already serving at least one of the cities of the 
new route, its "brand reputation," to the extent that such a thing 
exists, is already established on at least part of the route, so it will have 
an even easier time getting established than a totally new firm (this, 
however, is not very important if, in fact, the product is not very 
differentiable in the first place) . 

Overall, then, there is every reason to believe that entry barriers 
in the airline industry are low enough to allow it to function competi- 
tively in the absence of regulation. The issue of regulatory policy re- 
garding airline entry will be discussed further in the concluding sec- 
tion of this chapter. 

Cpmpetition and Product Quality 

If the evidence presented so far is correct, any efficiency arguments 
in favor of continued CAB regulation are weak ones indeed. However, 
one very important aspect of the problem has not yet been covered. 
The classical argument that competitive markets will generate an eco- 
nomically-efficient optimum goes as follows : if price is greater than 
the extra cost of producing a good, firms will have an incentive to 
produce more of the good, for they can continue to make money for 
each good made. Competition will drive the price down to cost, includ- 
ing a "normal" return, defined to be just enough to continue to induce 
investment in the industry. This argument, however, deals with a 
single, homogeneous product. We know that this is not the kind of 
product provided in the airline industry. Rather, there is a whole con- 
tinuum of different product qualities, and these qualities vary in sev- 
eral dimensions as well (seat size, meal service, and, most importantly, 
schedule delay time). 

1" Caves, p. 54. 


The question, then, is the following : will competition in price and 
product quality together result in an optimal outcome, an optimal com- 
bination of price and product quality ? Until recently, this issue had 
not been worked out theoretically for the most important aspect of air- 
line product quality, namely schedule delay. However, Gary J. Dor- 
man, in a recent doctoral dissertation under the direction of the present 
writer, has solved this problem, and shown that in the absence of any 
sort of scale economies or externalities, competition in the two dimen- 
sions of price and scheduling will indeed generate an optimum.^^^ The 
proof of this result is too elaborate to go into here, but the gist of the 
argument is simple : so long as a market is not providing an optimal 
combination of JPare and service quality, a new airline can enter the 
market and make money by approximating that combination more 
closely. Problems arise, however, if the amount at which different 
passengers value service quality differs considerably: then there is a 
conflict among passengers as regards desired service quality, and unless 
there are many flights, this conflict will be difficult to resolve. But if 
there is a conflict as regards value of service quality to customei^, then 
competition will generate a superior solution in all likelihood, how- 
regulation can do no bett4?r than competition ; the problem is inherent : 
ever, because it will probably enable a wider variety of prices and 
service qualities to suit varied tastes.^^^ 

From the. evidence presoiitod here. l)ot]i theoi-etiral a^id empirical, 
then, the upshot is straightforward: a loosening of trunk airline reg- 
ulation in the United States, with a greater reliance on market forces, 
would promote a greater degree of economic efficiency in the form of 
lower fares, and a service quality closer to what people are willing to 
pay for, than existing regulatory policies. Just how great the overall 
potential benefits of regulatory reform to domestic US air travelers is 
considered in th.e next subsection. 


An}^ estimate of the welfare loss or gain from a given policy is by 
necessity a rough one, because a number of simplifying assumptions are 
always necessary to make them. Nevertheless, as long as these estimates 
are interpreted and used with some care, they are considerably better 
than knowing nothing at all. The present section considers the welfare 
effects of CAB regulation for consumers, while tlie following section 
looks at those effects for producers. 

To estimate the economic effects of CAB regulation, we work in two 
steps: we first consider the likely impact of deregulation on fares, and 
second on service quality as measured by schedule delay. In calculatincr 
this loss, we restrict our analysis of coach and economy passengers. 
This is not a major shortcoming, since these classes of passengers now 
account for no less than ninety per cent of all scheduled domestic trunk 
passenger-miles.^-'^ We shall, however, consider qualitatively the im- 
pact of CAB regulation for first-class passengers, as well. We shall also 
restrict our analysis, as said in the beginning, to scheduled flights. 

"''Airline Corapptition : A Theoretical and Empirical Analysis (Ph. D. dissertation, 
Berkeley. 1976 ), chapter 3. 

"3 On low-density routes, this problem can be acute. However, because low-densitv 
routes tend to be short in haul, surface alternatives tend to allow some other price-quality 
alternatives here. 

i^U.S. Civil Aeronautics Board, Air Carrier Traffic Statistics (December 1975), p. 6. 

83-944—78 9 


The details of the calculations for the results presented here are 
shown in the Appendix to this study. The detailed calculations, as 
shown in the Appendix, are for the year 1974, perhaps the most recent 
year that CAB regulation continued in its fullest force, as previously 

Effects of CAB Regulation on Fare Levels 

To estimate the potential impact of deregulation on fares, we shall 
first calculate how fares are likely to change with deregulation for a 
"typical" trip in terms of length, number of stops, and aircraft type. 
We shall then aggregate to all trips by multiplying the likely fare 
reduction for a typical trip by the total number of such trips. If any- 
thing, this will understate the consumer gain from moving from exist- 
ing to optimal fares, because it ignores inefficiencies introduced by a 
nonoptimal fare structure of the sort mentioned in the previous subsec- 
tion. We shall attempt to get a rough notion as to how much additional 
cost is imposed by a nonoptimal fare structure later in this section. 

Another important assumption as relates to the estimated welfare 
loss is just what measure of actual, regulated fare should be used for 
the "typical" trip. The estimates of Douglas and Miller, Pustay, and 
others used actual revenue yields of CAB carriers. The main problem 
with it, as pointed out previously, is that it averages in many different 
discount fares with various trip restrictions. These fares provide serv- 
ice qualities different from the no-string-attached unregulated fare 
offered on intrastate routes in California and Texas. Thus, if a model 
predicting optimal (cost-based) fares on interstate routes also predicts 
accurately the unrestricted fares on intrastate routes, it is only reason- 
able that predicted efficient fares in interstate routes be compared with 
similarly unrestricted fares under CAB regulation. The importance 
of this is that using revenue yields as a measure of interstate fares (as 
Douglas and Miller have done) yields a downward-biased estimate of 1 
the welfare loss from CAB regulation. 

What we shall do, then, is to predict an "efficient" interstate fare 
for an average interstate trunk route, based on intrastate load factors 
and seating configurations. This fare is then compared with the actual, 
unrestricted CAB fare on such an average route. This difference be- 
tween the two fares is then multiplied by the total number of coach 
trips taken to get the full fare cost of CAB regulation to coach 

Although use of interstate revenue per passenger trip will yield a 
downward-biased estimate of the fare loss from CAB regulation, 
nevertheless, it would be useful to know what the welfare loss figure 
looks like based on the assumption that the regulated fare is the 
revenue yield. The reason is simply to get a "rock-bottom" estimate 
of the welfare cost of CAB regulation, and to test for the sensitivity 
of our results to variations in assumptions. Therefore, we present two 
estimates of the welfare loss to consumers from airline regulation, 
one based on each assumption regarding regulated fares. 

The calculations are based on the cost model presented in Keeler 
(1972), adjusted for 1974 prices as described in the Appendix. It was 
assumed, furthermore, that an unregulated carrier would use a 727-200 
aircraft for this typical route (the average trunk haul in 1974 was 795 
miles, a distance appropriate for this aircraft type).^^! \i y^^^ assumed 

121 Ibid., p. 6. 


also that an unregulated carrier could achieve a load factor of 60 per 
cent in an all-coach configuration. This figure reflects what the Cali- 
fornia and Texas carriers have achieved, and also the calculations of 
Douglas and Miller, which indicate that on a route such as this one, a 
load factor of 60 per cent is about right in terms of efficiency.^22 j3g. 
cause meal service would be required on a trip of this length, we allow 
for full galley space, consistent with a seating capacity of 153 passen- 
gers per aircraft. ^^^ 

Based on these and other assumptions discussed in the Appendix, 
an "unregulated-' market for this typical route would result in a fare 
of $46.70, or a yield of 5.88 cents per mile. For a typical route of this 
length, a regular fare of $67.60 was charged in 1972, for a revenue 
yield of 8.5 cents per mile. Multiplying this over all coach trips made 
'in 1974 yields a loss (excessive charge) to coach passengers of $2.7 
billion ; the fare loss so calculated is $20.90 per trip, or 31 per cent of 
the fare actually paid. 

The alternative calculation, based on the assumption that the regu- 
lated fare is the coach revenue yield, generates a lower welfare loss 
estimate. Specifically, the actual 1974 coach revenue yield was 6.94 
cents per passenger-'mile.^^^ Thus, the loss on the basis of this assump- 
tion is about 1 cent per passenger-mile, or $1.1 billion overall. By these 
calculations, the fare loss to coach passengers is about 15 per cent of 
the regulated fare paid.^-^ 
Effects of CAB Regiilafion on Coach Service Quality 

Let us now consider the service quality improvementvS brouglit on 
by CAB regulation. Again, we restrict our analysis to coach and 
economy service, for several reasons. First, there is strong evidence 
to indicate that first class service exists exchisively by grace of the 
cross-subsidy it currently receives from coach, which will be discussed 
in the next subsection; in other words, there is evidence to indicate 
that the vast majority of those who use first clnss serv^ice (at that now 
only ten percent of the total) would not be willing to use it if they had 
to pay the full long-run costs of providing it.^^*^ Furthermore (and 
along the same line of reasoning), no intrastate carrier provides first 

122 For route density of 400 passengers per day, Douglas and Miller find an optimal 
load factor (on a Boeing' 727-200) of fifty-seven percent with a $10 per hour time 
value, and sixty-one percent with a $5 per hour value. For a 1,000 mile haul, thoy find 
an optini.ll load factor of sixty to sixty-three percent, again depending on the value 
of schedule delay time. Thus, for the 792 mile haul discussed here sixty percent might 
seem roughly correct. It must be remembered, however, that because the operating cost 
data used here differ from those of Douglas and Miller, optimal load factor calculations 
based on them are likely to be different from those of Douglas and Miller, at least by a 
small amount. Thus, the' sixty percent assumed here Is an approximation. Dorman. op. cit.. 
provides a more detailed Integration of the present cost model with the Douglas and 
Miller stochastic delay model. 

1^3 The capacity of loo with full galley space is based on an estimate by Mr. Lawrence 
Guske of PSA as to how much the capacity of its 727-200's (which seat 158 with two 
coffee bars) would be reduced If they were provided with full galley space. Mr. Guske 
gave this estimate in a telephone conversation with Mr. Thomas E. Dooley of the U.S. 
General Accounting Office. 

i-*Air Carrier Traffic Statistics (December 1975), p. 6, and USCAB, Air Carrier Finan- 
cial Statistics (December 1975), p. 3. 

1^ That is (6.94— 5.,SS)/6.94=.15.*'.. In calculating this welfare loss, we have not attempted 
to calculate a "welfare triangle," which represents the extra benefits from extra air 
services consumed as a result of the lower fares, in addition to cost savings from existing 
consumption, for three reasons. P^irst, this welfare loss is likely to be quite small compared 
with cost savings. Second, it is highly dependent on the demand elasticity for air transpor- 
tation, and that we do not know with any certainty, as explained in the text. Third, for 
calculation of this value to be meaningful. It is necessary that prices of substitute goods 
(such as auto travel) reflect marginal costs as well. In fact, however, there is evidence 
that highway user charges for Intercity auto transport do not, in fact, reflect their costs. 
See Keeler. Resource Allocation in Intercity Passenger Transportation, chapter 4. 

1^ See the text below, under the subheading "The First-Class-Coach Differential." 


class service, so not only it is difficult to see a market justification for 
it, it is impossible to see what this service would be like (if it existed) 
in a deregulated environment. Finally, if first class service did dis- 
appear with deregulation, it is difficult to measure the full benefits 
which current first class users receive from the service which they 
would not receive in the absence of first class. 

To determine the likely effects of deregulation on trunk airline 
service quality, Ave make some further simplifying assumptions. As 
Douglas and Miller do for their welfare loss calculations, we assume 
that a typical route carries 400 passengers (in both directions) per 
day.^2^ In addition, we assume that each one per cent reduction in the 
coach fare level will induce an additional 1.2 percent increase in 
passenger travel on a route. This is consistent with previous demand 
studies, and it incorporates two effects : the effect on additional demand 
from a reduction in price, and the reduction in service quality which 
lower fares will induce (in other words, a lower fare induces the air- 
lines to achieve higher load factors in order to break even. This in- 
crease in load factors makes it harder to get reservations at some times, 
and may, as a result, drive away demand.). The dem.and elasticity 
assumed incorporates both effect, based on the most reliable available 
empirical evidence.^^^ 

On the basis of this assumption regarding route density, and with 
evidence on seating capacities of a typical coach compartment under 
CAB regulation, and of a coach-configured aircraft under deregula- 
tion, it is possible to calculate total schedule delay, before and after 
deregulation, confronted by the typical coach passenger (recall that 
schedule delay is the difference between the time the traveler vrould 
prefer to leave and the time he must leave, given flight frequency and 
seat availability on a given flight). The details for finding this sched- 
ule delay are given and discussed in Section lY of the Appendix. They 
are based on a functional relationship between aircraft size, passen- 
gers per aircraft, and route density developed and estimated by 
Douglas and Miller. Again, calculations were based on two alternative 
assumptions regarding the currently-regulated fare : the first is that 
the fare may most accurately be measured by the official, unrestricted 
fare on an "average route" (average in the sense of length) . The second 
assumption is that the typical fare is best measured by the revenue 
collected on such a typical route, averaging in the effects of various 
discount (but restricted) fares. 

The results of these calculations are surprising: they indicate that 
the typical coach passenger on regulated trunk airlines suffered as 
much or more schedule delay as a coach passenger in a deregulated 
environment would encounter. This is not so surprising as it may seem 
when one examines the load factor in coach service provided by the 
trunk carriers : in 1974 and 1975, it averaged around 58-59 per cent.^^^ 
Given that the "deregulated"' or "efficient" fares calculated here are 
based on a load factor of GO per cent (and that that is the load factor, 
in fact, achieved most often in recent years by the intrastate carriers), 
it is not surprising that coach passengers would find little, if any, 

'^'^ Economic Regulation of Domestic Air Transport, p. 172. 

^'^ This value of demand elasticity is the average of the Values estimated by the DeVany 
and Brown-Watklns studies previously cited. See Appendix footnote 31. 

I-'' U.S. Civil Aeronautics Board, Air Carrier Operating Statistics (December 1975). 
p. 6. 


deterioration in schedule delay with dereofiilation. And if one uses tlie 
official, unrestricted fare as the measure of the coach fare under CAB 
reirulation, the decline in fare with deregulation is so dramatic that 
there is enough induced demand from coach alone to actually increase 
flight frequency with deregulation, reducing schedule delay modestly, 
with a value of coach air travelers of over $200 million per year in 
schedule delay time saved.^^^ Again, details are shown in Section IV 
of the Appendix. 

On the basis of evidence from intrastate markets, the latter assump- 
tion about regulat-ed fares seems to produce results on flight frequency 
more consistent with reality: in these markets, whenever an intrastate 
carrier has moved onto a route, flight frequency has grown much more 
sharply than previously, and more rapidly than on equivalent routes 

The calculations on which these results are based are rough, and ob- 
viously not too much should be attached to the exact numbers shown 
in the Appendix as regards welfare losses and schedule delay times. 
Nevertheless, they indicate quite clearl}" that if a 60 per cent load factor 
is representative of the likely outcome on a typical deregulated route 
(and evidence indicates it is), then the much-touted service benefits 
which CAB regulation confers on coach passengers are quite negli- 
gible, and may even be negative. 

Does this mean that CAB regulation has conferred no benefits of 
increased service on the traveling public? Certainly not. They imply 
instead that practically all the service l^enefits of the regulation have 
accrued Ir) the teii pei- cent who I'ido in first class.^ - Fii'st class load 
factors are far lower than coach (only 39.7 per cent in 1974 and 34.7 per 
cent in 1975), and first class passengers receive numerous other service 
amenities not received in coach. 

In conclusion, then, we can assert with reasonable confidence that the 
welfare loss from CAB regulation to trunk coach passengers in the 
United States in 1974 was ?^l-2.7 billion, excluding the costs of a non- 
optimal fare structure, and excluding losses from inefficient invest- 
ments in technical progress induced by regulation. These costs will be 
investigated in the following two subsections of the present section. 

Meanwhile, it is worth repeating that the loss calculated so far is 
not totally deadweight in nature : an unknown fraction is redistributed 
in benefits to the privileged ten per cent who flv in first class. Although 
these benefits cannot be measured, more should be said about the cur- 
rent benefits of regulation to first class passengers, and that is done in 
the following subsection, which is on the costs of CAB regulation in 
nonoptimal fare structure. 

Effects of Regulation on the Structure of Fares and Service Quality 

There are two types of inefficiency in fare structure which, it can be 

argued, have been brought about bv CAB regulation: the first is an 

inefficient differential between coach and first class fares. The second 

""This assnmps a ?5 per hour Aaliie of sfhpdnip dplnv timo. Tf is diffiniU to belipve 
that the avprare popch air travplpr in a dprpcnlatpd market wonld have a scheduled dplav 
time value hlchpr than this, for reasons jrlven in the text in the discussion of schedule 
delar time valuation, above. 

1^ Kennedy Report, pp. 40-48. The recent introduction of sharplv lower fares hv South- 
west Airlines has had an especially strong impact on both traffic and flieht frenuencv 
in the Rio Grande Valley in Texas. Since Southwest entered the market, the number of 
fli£rht« servinnr Harlincen. Tex. has risen from 37 to 61 (Kennedv Report, p. 47). 

1-2 This is discussed in more detail in "The First-Class-Coach Differential," on p. 122. 


is relationship and fare and fare and distance, which in both coach and 
first class does not necessarily reflect the optimal relationship. 

The First Class-Coach Differential 

There is considerable evidence that first class service is cross-sub- 
sidized by coach, and that this has gone on for some years. For 1968, 
for example, the present writer found that while a first-class seat took 
up 1.68 times as much space as a coach seat, first class fares yielded 
only 1.36 times as much as coach fares. Furthermore, in the same year, 
the coach load factor was 56 per cent, compared with only 43 per cent 
in first-class. Overall, the revenue yield of a first-class-seat-mile was 
2.98 cents, whereas the yield of 1.68 coach seats (one first-class space 
equivalent) was 4.80 cents per mile.^^^ It is this sort of cross-subsidy 
from coach to first class which seems to be responsible for a good part 
of the high price-cost margins for coach passengers observed in the 
previous section, with no corresponding service improvements to those 
passengers in the coach compartment. On the basis of this evidence, it 
can be shown that the first-class fares collected in the late 1960's and 
early 19T0's were lower than the lons^-run costs of first-doss servioe.^^* 

Why should the airlines favor the continuation of this sort of fare 
differential, given that it does not seem profitable in the long run? No 
certain answer can be given, but the following one is reasonable and 
consistent with the statements of various airline executives. ^^^ The air- 
lines have excess capacity, and they have had it for a long time. In 
the short run, it will increase profits to use a discriminatory pricing 
scheme which extracts extra payments from those willing to pay extra 
for the space which this excess capacity allows. Even if the extra fare 
charged for this capacity does not totally cover its long-run costs, it 
will be profitable in the short run, if the airlines are stuck with the 
capacity. Thus, the airlines tend to move the bulkhead between first 
class and coach so as to keep the coach compartment relatively crowded, 
while the roomier seats in first-class stay relatively empty, but the air- 
lines do extract extra payments from those willing to pay something 
extra for the benefits of space and easy reservations during peak 
periods, plus other amenities such as finer m.eals and complementary 

Although these policies may make sense as a short-run profit-maxi- 
mizing strategy on the part of the airlines, they make less sense in the 
long run; indeed, if this story is true, and if airlines did maximize 
profits, it is difficult to see why an airline would configure new aircraft 
to accommodate first-class passengers, unless it perpetually overesti- 
mated its capacity needs and then price-discriminated to take advan- 
tage of that. 

This problem of the first-class-coach differential was one which was 
considered by the CAB during the DPFI. After some vacillation on 
the issue, the CAB decided (in 1974) that fares should reflect costs 
in both compartments (based on a 55 per cent load factor), and that 
the cross-subsidization should thus be eliminated, but gradually over 
a period of several years. ^^^ As a result firpt-class fares have been 

i33Keeler, "Airline Regulation and Market Performance." p. 419. 

"*For a rough calculation of the first-class loss for 1974, see the Appendix, Section V. 

is-'^This is also the explanation given by Douglas and Miller, pp. 102-103. 

1^ One fare increase came Apr. 29, 1975, and another Apr. 1, 1976. The final increase 
occurred on Apr. 1, 1977. See "Seat Density Changes May Signal Trend." Aviation Week 
104 (Jan. 19, 1976), pp. 25-26. See also A. E. Kahn. interview, in Congressional Record — 
Senate, Feb. 21, 1978, p. S. 1978. 


increased more sharply than coach over the past three years, with 
the result that demand for first-class service has shrunken sharply, 
while at the same time the airlines have been reducing the size of their 
first-class compartments."" This supports the arguments made earlier 
that if (or when) first-class fares reflect full long-run first-class costs, 
the demand for first-class service will shrink to nearly zero. 

Meanwhile, reduction of the cross-subsidy between first class and 
coach has narrowed the margin between actual and optimal coach 
fares. Thus, as previously stated, the markup of actual interstate 
coach fare over hypothetical unregulated fare on twenty-eight top 
interstate routes declined from 20-90 per cent in 1968 to 30-56 per cent 
in 1974. Nevertheless, a markup of 30-56 per cent is not a trivial one, 
so that the reduction of this cross-subsidy which occurred between 
1968 and 1974 did not come near eliminating the welfare cost of regu- 
lation to coach passengers. Furthermore, recent work by Pustay has 
shown that even if the cross-subsidy were eliminated completely, there 
would still be considerable welfare cost from nonoptimal fare struc- 
ture due to CAB regulation. It is to his results which we now turn. 

Costs of a Non-Optimal Fare Structure hy Route Type 
Pustay's analysis of the costs of CAB regulation is more accurate 
than those of previously mentioned studies, in that he calculates 
optimal fares and service qualities by route type (based on the Doug- 
las-Miller schedule delay function), compares actual and optimal 
fares and service qualities on each route type, and adds up welfare 
losses over all routes. Thus, he estimates the cost of a nonoptimal fare 
structure over a wide range of routes, as well as the costs of a non- 
optimal fare level. T'nfortunately, Pustay's welfare loss estimates are 
downward-biased relative to tliose presented in the subsection above, 
for in these calculations he has ignored the costs of an inefficient differ- 
ential between first-class and coach service, and has also used the 
CAB's long-run cost model, which probably overstates the operating 
costs (at a given load factor) which a carrier of the intrastate type 
could achieve. ^'^^ (Pustay does acknowledge, indeed, asserts strongly, 
that CAB regulation extracts higher costs than his estimates because 
of these two factors — he devotes two chapters of his thesis to them ; but 
he does not quantify these costs). Nevertheless, Pustay estimates that 
as of 1969, CAB regulation extracted a welfare cost of $660 million 
at a So per hour value of schedule delay time, and $490 million with a 
$10 value. 

Ideally, it would be desirable to integrate the cost model used for 
welfare cost calculations earlier in this section with route-by-route 
demand data, to get a more accurate estimate of the welfare cost of 
regulation. However, sufficient data and resources were not available 
at the time of the writing to make this possible. 


It is agreed by most that the airline industry operates subject to 
constant returns to scale. Furthermore, it earns normal profits, and 
appears that it would earn normal profits in a deregulated world, as 

"'Ihid., pn. 25-20. 

^^'The Effects of Regulation on Resource Allocation in the Domestic Trunk Airline 
Industrv, oh. 4. 


well. It is tempting to argue, then, that "in the long run" there would 
be no loss to producers in the airline industry from deregulation. But 
this is not a satisfying conclusion, for two reasons : first, there could 
be considerable short-run losses of profits moving from the regulated 
equilibrium to the deregulated one; furthermore, the demand for 
labor in the industry could shift substantially, increasing or reducing 
the nutnber of jobs in the industry. Secondly, even if firms were no 
better or worse off from deregulation, some might be helped, and 
others hurt. In other words, the transition period to a deregulated 
environment could involve two types of gains or losses : overall short- 
term gains or losses to specific factors or production in the industry, 
and short-term gains and losses to specific firms (or factors hired to 
specific firms ) . We consider both. 

It can be argued that the overall gains and losses to both capital and 
labor in the industry resulting from deregulation are likely to be tied 
quite closely to the relationship between fares and the derived demand 
for aircraft capacity. In other words, if a decrease in fares tends to 
increase the total number of flights supplied by the industry in the 
US, that will tend to increase the return both to capital and labor : 
since aircraft capacity is the most important type of capital, an in- 
crease in the number of flights will increase the demand for aircraft. 
This will benefit both holders of existing aircraft and producers of 
new ones. Similarly, it will benefit flight crews. The opposite will hold 
if deregulation reduced the demand for aircraft capacity. 

Now, it has been asserted by Douglas and Miller that existing CAB 
policies have set fares at a level so as to maximize the number of seats 
provided for air trips in the United States, and that the higher load 
factors resulting from deregulation would reduce the number of 
flights and the derived demand for aircraft capacity .^^^ 

However, the evidence from intrastate routes in California and 
Texas suggests a very different story. In nearly every case studied, 
the total number of flights grew after the institution of low-fare 
service provided by intrastate carriers. (The load factor rose because 
the number of passengers rose more rapidly than the number of 
flights, and on the basis of evidence from previous studies, it is diffi- 
cult to find anything which would cause traffic to grow so rapidly on 
these routes save for the lower fares.^^° Thus, the few instances we 
have seen of dramatic fare reductions would seem to contradict 
Douglas and Miller's assertion that a reduction in fares would reduce 
the demand for aircraft capacity. 

The Appendix (Section I) challenges the assertion that CAB .fares 
are currently set to maximize the number of flights. The details are 
too technical to go into here, but the basic idea is that this result 
depends crucially on the empirical value of a specific demand param- 
eter (the amount by which a one per cent fare cut would increase 
traffic on a typical route). We simply do not have sufficiently accurate 
estimates of this parameter for the trunk air routes of the US to know 
the impact of a fare cut on the amount of capacity provided. 

Nevertheless, the method presented earlier in this section, nnd de- 
scribed in detail in the Appendix, can be used to predict the impact 
of deregulation on the number of flights offered. As can be seen from 

^39 Economic Rejmlation of Domestic Air Transport, p. 176. 
1^0 See Kennedy Report, pp. 46-48. 


Table I of the Appendix, the result is ambicruous, and depends on 
what one assumes to be an accurate measure of the fare under current 
CAB regulation. If the revenue yield is an accurate measure of the 
fare, then the number of flights would fall with deregulation. If, on 
the other hand, the unrestricted fare is the accurate measure, the num- 
ber of flights would go up. The basic question is whether the flight- 
increasing tendenc}' of an increase in demand resulting from a lower, 
less-regulated fare will offset the flight-reducing effect of an increase 
in load factor. Although the results of the theoretical model are am- 
biguous, the actual results from deregulated routes indicate that the 
number of flights could easily rise from deregulation. 

The roughness of these results is obvious. However, what will hap- 
pen on any one route is even more ambiguous: the few studies of 
demand on individual routes which have been made (by observing 
changes in fares and traffic over time and trying to net out other 
factors) would seem to indicate that the fare elasticity of demand 
(and, hence, the extra traffic which a fare cut would induce) varies 
widely from route to route, depending on numerous factors, including 
available surface alternative transportation, percentage of traffic 
made up of business versus pleasure travelers, the attractiveness of 
the route for conventions or vacations, etc.^^^ 

As a result of this, it is likely that deregulation would increase the 
demand for flights on some routes, while decreasing it on others. This 
means that if a carrier served routes with mainly low demand elas- 
ticities, it could, indeed, suffer losses from deregulation, although 
it is difficult to predict specific carriers which would lose. Other 
carriers would gain. 

Route structure, however, is not the only reason why some carriers 
could suffer losses (and lay off workers) as a result of deregulation. 
In any competitive market, it is to be expected that new firms will 
attempt to enter, and not all will survive in the long run. CAB reg- 
ulation has suppressed both the entry and the exit which is necessary 
for these market forces to function. Thus, it would not be surprising 
if. under deregulation, a few of the less efficient firms were competed 
out of business by more efficient firms. One would expect this process 
to £ro on continually, renewed by firm entr^'. 

The point of all this is that while deregulation could benefit holders 
of capital invested in the airline industry, it is likely that some air- 
lines, some investoi^. and some workers will lose in the short run. 


Becau'=e the trunk airlines are subject to entry regulation, and be- 
cause they are free to seek fare increases if their profit levels are 
'•inadequate,-' one might be inclined to believe that managerial ineffi- 
ciencies could creep into their operations, might fail to minimize 
costs for the outputs they produce. 

"^Miether such inefficiencies do or do not exist in the airline industry 
is a matter of some controversy, and. before presentinc: 'iny conclu- 
sions on this matter, it is worth briefly summarizing alternative view- 
points on this matter. 

"1 Spp. for pxamplp. Philip K. Vprleprer. Jr.. "Models of Demand for AJr Transportation." 
Bell Journal of Economics 3 (Autumn, 1972), pp. 437-457. 


Let us first consider those arguments, theoretical and empirical, 
which support the notion that the airline industry should possess 
managerial inefficiencies. This argument most logically stems from the 
notion of "sheltering" from competition mentioned above, and has 
been supported directly with some empirical evidence. For example, 
Eobert J. Gordon, some time ago, found that trunk air carriers with 
more "competitive" route structures (i.e., more service rivals on their 
routes) had lower costs than those carriers with fewer rivals.^*- More 
recently, evidence presented by William Jordan has indicated that 
the California intrastate carriers achieve higher levels of productivity 
in all their cost accounts than do interstate carriers, and that even 
their higher load factors may be due partly to better scheduling prac- 
tices, i.e., more planes at peak periods than the trunk carriers sched- 
uled*^ Finally, theoretical economic models have been developed by 
notable economists (Oliver Williamson, Richard Cyert, and Herbert 
Simon, to name a few) describing why, in a sheltered market environ- 
ment, one would expect managers to pursue goals other than maximum 
profits, and why, as a result, inefficiencies could creep into firms.^'** 
There is some evidence, then, both theoretical and empirical, to make 
us believe that trunk airline managements are indeed inefficient as a 
result of the sheltered environment afforded by the CAB. 

However, there is also evidence to make us doubt that managerial 
inefficiency is a serious source of welfare loss from CAB regulation 
(above and beyond the losses stemming from service competition of 
various sorts, which is socially inefficient, but not from the viewpoint 
of a non-colluding, profit-maximizing firm) . 

First, and more importantly, although airlines do not compete in 
price, they do compete in service quality. Thus, on a given route, if one 
airline is less efficient than another, it will not be able to provide so 
good a service quality as others for a given fare level (this can involve 
any of the various dimensions, of service quality discussed previously 
in this section). There is every reason to believe that the com.petitive 
pressures to provide high service qualities are every bit as great in the 
trunk airline market as are the pressures to operate at the lowest pos- 
sible price in a price-competitive market. 

There is also more direct empirical evidence that casts doubt on the 
hypothesis that a more sheltered regulated environment causes mana- 
gerial inefficiency. Using data for the late 1960's Douglas and Miller 
attempted to redo the Gordon study comparing airline costs with the 
number of "competitive" routes each airline faced.^*^ The Douglas- 
Miller results indicated no meaningful relationship, unlike the results 
of the earlier Gordon study. And Douc^las and Miller's statistical tech- 
niques were more sophisticated than Gordon's. 

As regards the intrastate carriers, it is difficult to separate com- 
pletely the extent to which their lower costs stem from a different serv- 
ice quality relative to the trunk carriers and to what extent they stem 
from superior managerial efficiency. It will be recalled that the 1972 

142 "Airline Costs and Manajrerlal Efficiency." 

143 Airline Regulations In America, chapter 11, 

"* See, for example, H. Simon, "Theories of Decision-Making in Economics and 
Behavioral Science." American Economic Seview 49 (June 1959). pp 253-283- O E 
Williamson, "Managerial Discretion and Business Behavior." American Economic Review 
53 (December 19fi3) ; pp. 1032-1057; and R. M. Cyert and J. G. March, A Behavioral 
Theory of the Firm (Englewood Cliffs, N.J., Prentice-Hall, 1963) 

1*5 Economic Regulation of Domestic Air Transport, pp. 141-144 


stiidv done bv the present writer (based on IOCS data) used a cost 
model estimated for the trmik airlines, but that model accurately pre- 
dicted fares on the California intrastate routes. It is true that this 
model did attempt to correct for some aspects of perceived managerial 
inefficiency on the trunk carriers,^'^ and that may be why it managed to 
predict California intrastate fares using cost data for the trunks. How- 
ever, it would appear from this work that a very large part of the dif- 
ference between trunk and intrastate costs can be explained by such 
basic differences as seating configurations and load factors. 

Thus, while there is some evidence that CAB regulation has allowed 
managerial inefficiencies to creep into the trunk airline industry, that 
evidence is anything but conclusive. 


So far, our discussion ha? taken technology as given. That is. it has 
assumed that the "state of the art'' in aircraft, ground equipment, 
pilot training, and firm organization are given, and are unaffected by 
regulatory policies. However, over the long run, regulation can affect 
the very develo]Dment of technology. Although the costs and benefits 
of regulation as it relates to technological change cannot be discussed 
in the same quantitative context as the static costs and benefits (i.e., 
those calculated with technology' assumed to be given), nevertheless, 
it is worth saying something about the impact of CAB regulation on 
technological progress in the airline industry. 

Because CAB regulation lias tended to hold fares at an artificially 
high level, forcing competition to occur in the form of product rivalry, 
one would a priori expect part of the rivalry to take the form of 
product sophistication. That is to say, the airlines might be expected 
to invest in more advanced and sophisticated equipment as a result 
of CAB price regulation than they otherwise would do. 

Quantitative evidence on this is difficult to find, but it is possible 
to tell stories which illustrate that this is indeed true. Perhaps the 
most famous and striking such story is that of the Douglas DC-7, 
first told in detail by Aaron Gellman.^"' It will be recalled that es- 
pecially during the early and mid-lOHO's, air fares were set to yield 
much higher profits on long hauls than on short hauls. The net impact 
of this was more frequency competition and lower load factors on 
long hauls than on sort hauls. But competition took a different twist 
here in the 10r)0*s: before the DC-7 there had been no aircraft with 
a transcontinental, nonstop capability — at least none which were any- 
where near so economical in operation as the Douglas DC-6B, which 
required a refueling stop to cross the continent. Furthermore, it was 
known throughout most of the lOoO's that by the end of the decade, 
commercial jet aircraft would be available, making for lower oper- 
ating costs on long hauls than even the DC-6B. Xevertheless. prices 

^♦^ SpecificaHy, the 1968 calculations selected coefficients for the nationwide airline with 
the lowest indirect operating costs. American In that year. Furthermore, the coefficients 
of the indirect cost equation mny have been downward-biased in another way. as well : all 
passenper-mile ficrures were converted to ton-mile figures and added to freight ton-mile 
figures for the ton-mile variable. Because the coefficient for ton-miles reflects both passen- 
ger and freight operations, and given that freight operations are probably cheaper to 
provide than passenger operations in Indirect costs, this biases downwards the coefficient 
if it is used to predict passenger costs, as It was in mv paper. 

1'" Aaron J. Gellman. The ECFect of Reirnlation of Aircraft Choice (Ph. D. dissertation, 
Massachusetts Institute of Technology, 1968). 


were set sufficiently above costs on transcontinental runs to make 
service rivalry using newly-developed nonstop propeller aircraft via- 
ble from a, carrier point of view. Thus, the DC-T was developed by 
Douglas and bought by several transcontinental airlines during the 
1950's. despite the fact that its seat-mile costs were higher than those 
of a DC-6B, and despite the fact that the DC-7 was rendered obso- 
lete by jets within only five years or so of its introduction. Gellman 
showed that without the high transcontinental fares resulting from 
CAB regulation, the DC-7 would never have been developed, for its 
benefits were far outweighed by its costs from a social viewpoint. 

It is possible to tell other, less-striking stores about the impact of 
CAB regulation on air transport technological development, but they 
are more ambiguous in the sense that it is more difficult to determine 
whether an unregulated market would have generated similar tech- 
nological outcomes.^*^ The story of the DC-7, in any event, gives a 
good indication of what can happen to technological progress in a 
regulated environment. 

Goals and Achie\'ements of Trunk Airline Regulation — 
An Evaluation 

At this point, it is appropriate to review the goals of the Civil 
Aeronautics Act of 1938, along with the practical goals which the CAB 
has inferred from them, and to attempt to evaluate the extent to which 
CAB policies have been successful in achieving these goals. Rather 
than requote the goals stated in the 1938 Act, we simply list the goals 
in it, or distill from the discussion of the first section of this chapter. 

economic EmCIENCY 

Although economic efficiency may not be at the top of the 1938 Act's 
priorities, it is there, in its prescription for "economical, efficient serv- 
ice by air carriers at reasonable charges." The previous section pre- 
sents considerable evidence that the goals of economic efficiency in 
trunk airline service would best be promoted by loosening or eliminat- 
ing the present regulation of the CAB. If that has been a high-priority 
goal, it has nevertheless not been achieved. Airlines may be efficiently 
operated given their regulatory constraints, but, as we have shown, 
these constraints have imposed social inefficiencies on the market. 


The 1938 Act calls for "the encouagement and development of an 
air transportation system properly adapted to the present and future 
needs of the foreign and domestic commerce of the United States, of 
the Postal Service, and of the National Defense." ^*^ As previously 
stated, the CAB has interpreted this and other portions of the Act as 
implying that it should attempt to provide a complete, nationwide 
airline system with a minimum of direct subsidy. Hence, a system of 

"8 Jordan discusses product Improvements in chapter 3 of Airline Regulation in America. 
As to whether the inter- or intrastate carriers moved faster on this count, he concludes 
that the interstate carriers tend to introduce innovations somewhat faster. But the resutls 
were rather ambiguous. 

i« Sec. 102. 72 Stat. 740. 49 USCA 1030. 


price discrimination designed to cross-subsidize uneconomic routes 
with revenues from economic ones has developed. Given the uproar 
which often occurs with the abandonment of any transportation serv- 
ice, be it ever so uneconomic, and given that no Congress likes to spend 
more on subsidies than it must, this is not an unreasonable mterpreta- 
tion of the law on the part of the CAB. 

In the 1930's, when the vast majority of airline routes were not eco- 
nomically viable, it is possible to see some rationale for attempting to 
minimize direct subsidies through a discrimmatory pricing scheme. 
But things have changed since the 1930's ; the trunk air carriers have 
been off direct subsidies for nearly twenty-five years. Local service 
airlines have come into existence with direct subsidy support, with the 
specific purpose of serving small communities, and a third level of 
carriers, commuters, have come in to serve routes which even the local 
service carriers find unprofitable with subsidies. It is difficult to see the 
point of using profits from one route to support losses on another. 

The overwhelming amount of evidence indicates that, except for the 
cross subsidy from first class to coach, the CAB has managed to achieve 
very little cross-subsidies anyway : the potential profits on high-density 
routes tend to be competed away through service competition, leaving 
little to support the low-density routes ; trunk carriers have as a result 
tended to abandon low-density routes. That this is the case was first 
documented in the early 1960's by CaA'es.^^° 

There is additional evidence supporting the hypothesis that cross- 
subsidies are basically nonexistent among the trunk carriers : they have 
no incentive to provide them. If an airline operates some routes at a 
profit and some at a loss, any rational manager will attempt to aban- 
don (or convert to a profitable basis) the unprofitable route. The CAB 
has been quite liberal in allowing carriers to abandon unprofitable 
routes, and even if a carrier is forced to serve a given community, it 
has great flexibility in its ability to cut down service to a bare bones 
level.^^^ It is unclear why any airline should claim (as some have 
done) to be throwing away stockholders' money operating unprofitable 
service which it does not have to provide. 

On close examination, routes which trunk carriers claim ''unprofit- 
able" (but which they do not petition to abandon) are less unprofitable 
than they might seem. A recent example of this is provided by evidence 
presented to the Kennedy Subcommittee by United Airlines, document- 
ing alleged cross-subsidies.^^^ At first glance, fully one-half of the 872 
city pair routes served by United appeared to be unprofitable. But 
that analysis was based on "fully-distributed" costs, allocating in por- 
tions of such items as the cost of mowing the headquarters lawn into 
the costs of every route. On the basis of incremental (avoidable) costs, 
only 58 of the routes were unprofitable. (If an airline has some equip- 
ment available which would otherwise be unused, it is quite reason- 
able that it should use it for marginal services as long as these services 
cover more than avoidable costs.) Of these, four routes were flown to 
reposition aircraft, 17 were flown because they generated traffic which 

150 See discussion undor "CAB Policies in Administerin;? the Civil Aeronautics Act of 

151 cjpp Coleman, pp. ST-IO.*?. See also T'.S. Department of Transportation. Comment-* 
on the Answer of the Association of Locn] Transport Airlines to the Report "Air Service 
to Small Communities." processed, Washington, July 19, 1976. 

1"- Kennedy Report, p. 67. 


was profitable once it flowed into the network, and eight were on routes 
below 60 miles in length (so one could question the need for trunk 
service there). So only the remaining 29 routes might be abandoned 
and cause any potential loss to communities in the event of deregula- 
tion. These routes accounted for only .5 per cent of United's revenue 
passengqr-miles, and the total loss was only $5.5 million, or .3 per cent 
of United's total revenues. 

Thus, the evidence of United provides little or no evidence of cross- 
subsidization. Indeed, the only study showing significant cross-subsi- 
dization in recent years was one commissioned by the Air Transport 
Association (1975). It found that with deregulation, there would be 
372 city pairs in the U.S. which would lose service, and that service 
would be cut back sharply for nearly all remaining city pairs.^^^ 

This study has numerous problems. First, like United's initial sub- 
mission to the Kennedy Subcommittee, the ATA report takes no ac- 
count of the fact that some of the 372 routes do not have service even 
with regulation ; a good portion of the routes are profitable for their 
feeder value; a number are used to reposition aircraft; and a number 
of the routes are under 60 miles in length.^^* Also, the ATA's statisti- 
cal approach appears to bias its results toward unprofitability of a 
given route : there are all of ten city-pair routes within California and 
Texas, which it finds would lose service without CAB regulation, but 
which, in fact, receive service from non-CAB-regulated, non-subsi- 
dized intrastate carriers.^^^ These carriers earn substantial profits, as 
previously stated. This is straightforward, direct evidence contradict- 
ing the conclusions of the report. Finally, the ATA study's result that 
many currently profitable routes would sustain great service cutbacks 
in the event of deregulation stems from a bizarre and unrealistic as- 
sumption of the study: that with deregulation, all service would be 
taken over by a single monopolist, who would hold fares at their cur- 
rent levels and then cut back service until total industry profits were 
maximized, resulting in an 83 per cent load factor.^^^ This is implausi- 
ble because CAB regulation now is concerned exclusively with fares 
and entry of firms, and not with service quality. Yet the ATA study 
implies that with deregulation of fares and entry, the airlines will 
keep fares the same, while at the same time in some mysterious way 
colluding perfectly to reduce service competition and quality. There 
is, in fact, nothing connected with deregulation of fares and firm en- 
try which will make it easier for the airlines to coordinate service 
quality (so as to maximize joint profits) than it is for them to do so 

When all things are considered, the ATA study represents an inter- 
esting (if imperfect) exercise to determine what would happen if all 
trunk air service in the US were taken over by a monopolist, with com- 
plete and total barriers to entry. But it has absolutely nothing to do 
with what would happen with deregulation (and hence more open 
entry of firms) of American air transport. 

When all is said and done, how much damage is likely to be sus- 
tained by various communities from loss or air service stemming from 

153 Air Transport Association, Consequences of Deregulation of the Scheduled Air 
Transport Industry (processed, Air Transport Association, Washington, April 1975). 

154 Kennedy Report, p. 67. 

155 Consequences of Deregulation of the Scheduled Air Transport Industry, Exhibit B. 
^ Ibid., figure 4, 


dere^lation of trunk airline?? Extrapolating on the basis of the 
United report, the Kennedy Subcommittee estimated total subsidy 
needs to preserve service for all communities losine: it at under $25 
million per year.^^^ This is very small, indeed, compared with the 
$1-2.7 billion or larger welfare loss from regulation cited above. 

If there is any cross-subsidization resulting from CAB policies, it is 
from coach passengers to first-class. It is difficult to see what social 
purpose is served by this, especially given its regressiveness. Indeed, 
the CAB has itself resolved to eliminate this, after twenty-five years 
of vacillation on the issue.^^^ 


The needs of the post office and the national defense get important 
play in the Civil Aeronautics Act of 1938. Yet exactly what CAB reg- 
ulation has done to furtlier their needs is very much in question. 

Consider the case of the postal service. Levine (1975) has recently 
concluded detailed and exhaustive study on the impact of CAB (and 
previous) airline regulation on the postal service. LeWne concludes 
that throughout the entire history of airline regulation, the effect of 
regulation has been to achieve "mail rates . . [which] reflect an effort 
to give the carriers revenues in excess of those which they would re- 
ceive is prices reflected marginal costs." ^^^ 

The most recent (and important, from a current policy viewpoint) 
evidence of this presented by Levine relates to the legislative activity 
leading to the Postal Reorganization Act of 1970. More specificallv, 
in order to reduce its costs, the Post Office wanted to have the option of 
either CAB-regulated rates (as before), or voluntarily negotiated 
rates, handled individually by the Post Office for each carrier. Al- 
though the Post Office favored this reform to cut its costs, both the 
carriers and the CAB opposed it vigorously, on the grounds that it 
would reduce airline postal revenues below their current levels. In the 
end, this reform, though it had been in practice for some vears in the 
case of railroads, was not passed by Congress. To the dissatisfaction 
of the Postal Service, the CAB still rejrulates mail rates. Levine cites 
other evidence as well that in dealings with the Postal Service, the 
CAB has favored rates higher than the market would set.^«° Further- 
niore, Levme's study finds no evidence that CAB regulation has been 
m any way beneficial to the needs of the Postal Service. It thus ap- 
pears that, contrary to the Act, CAB policy does not serve the needs of 
the Postal Service. 

Before discussing the current role of CAB policies in the interests 
ot national defense, it is worth recalling an important fact about the 
industry in the 1930's, when legislation for current regulatory policies 
was passed: at that time, the industry was not economically viable 
without some sort of government support, at least on the vast maioritv 
ot routes. Thus, if the industry was to exist, the government would 
nave to sup port it economically, and an important part of the CAB's 

^5" Kennedy Report, p. 67. 

^ See the previous section. 

i^Reffulatinc Airmail Transportation," pp. 357. 


rolo was administoring this support. Perhaps one important reason for 
supporting- an industry which was not economically viable was to pro- 
Aade a pool of aircraft and trained pilots, ready to come to the aid of 
the nation in one way or another in case of a war. In this sense, it can 
be said that CAB policies in the 1930's were in the interest of further- 
ing the national defense. 

The situation, however, has changed dramatically since the 1930's. 
The trunk airline industry is overall quite viable economically now, 
and has been ever since the early 1930's.^^^ Given that this has been the 
case, it is difficult to point out anything which the CAB has done since 
World War II to further the interests of national defense through 
airline regulation. 

As regards the needs of the military, there is little evidence of any- 
thing specific which the CAB has done for its specific benefit; there 
are a few cases where it has authorized greater-than-normal subsidies 
for local ser^dce carriers to serve communities with military bases, but 
that is about all.^^^ Another argument regarding the national defense 
is that the policies of the CAB have tended to bring about more capac- 
ity than would be available otherwise, and that this extra capacity is 
potentially useful for military emergencies. We have already seen, 
however, that it is questionable that CAB policies have encouraged 
the building of more aircraft capacity than would otherwise exist ; the 
contrary is inore likely to be true. Futhermore, whereas existing regu- 
latory policies have brought about inefficiently-configured aircraft 
(low seating densities, first-class, lounges), not the best sort for a 
mobilization. Our evidence would seem to indicate that a less regulated 
industry would entail considerably more efficient capacity use — higher 
seating densities and less first-class, and hence, capacity more useful 
to the military. 


The 1938 Act charges the CAB with promoting sound economic con- 
ditions among the carriers. The evidence we have presented implies 
that the carriers have done no better or worse than they would have 
done in competitive markets profit-wise, and intrastate carriers outside 
the CAB's jurisdiction, at least the ones successful enough to survive, 
have done at least as well. Thus, while the CAB may have done the 
airlines little harm financially, there is little evidence that it has done 
them much good, either (save, perhaps, for preserving inefficient car- 
riers). Overall, there is little reason to believe that CAB regulation is 
needed for efficient firms in the industry to earn normal profits, and 
without regulation, only those firms which provided the most efficient 
service would survive. There is, in short, reason to question both the 
goal as set forth in the Act and the CAB's administration of the law 
in pursuit of the law. 

i«i It must be pointed out that aUhough the trunk carriers went off subsidies as of 
1952, Caves found much evidence of indirect subsidies for airports, airways, and the like 
into the late 1950's. See Air Transport and Its Regulators, pp. 411-418. On the other hand, 
the present writer found that with the 8 percent ticket tax, the amount of such subsidies 
as of the late 1960's was very low, if it existed at all on higher-density trunk routes. 
See Resource Allocation in Intercity Passenger Transportation, pp. 115-136. 

lea^j. Transport and Its Regulators, p. 434. 


destructht: competition 

Preventing destructive competition is another goal with vrhich the 
CAB has been charged. We have already seen that the destructive com- 
petition which occurred in the industry in the 1930's was a result of 
three things : the fact that the industry was not economically viable 
without subsidies at the time, the Depression, and the specific bidding 
system used by the Post Office to give out airmail subsidies in the mid- 
1930*s. Xone of these factors are relevant to the current situation, and 
we have, in fact, argued in the previous section that except under rare 
circumstances, destructive competition is not likely to occur in the 
trunk airline industr}-. Hence, the need for the regulatory goal of 
preventing destructive competition is highly questionable. 


Safety is another important goal of the 1938 Act ; but it is important 
to note that safety reguhation is now the domain of the Federal Avia- 
tion Administration, and not the Civil Aeronautics Board, and that 
no one proposes any changes in the FAA's enforcement standards. 

Xevertheless, it has been argued that price competition on the part 
of airlines will induce them to "skimp" on expenditures necessary for 
safety, and that as a result passenger safety will suffer under deregula- 
tion. There are several reasons to question this assertion. 

First, if a carrier is found to be in financial difficulty, the FAA 
increases its surveillance of the carriers.^^^ For this and perhaps other 
reasons, empirical studies indicate no correlation between a carrier's 
profitability and its fatalities.^^* 

Se<:'ond. it is not clear that an environment of price competition will 
give carriers any incentive to "cut corners'" on safety. Certainly, if 
word got out that a carrier was significantly less safe than other'car- 
riers. that would put it out of business rapidly. Second, not only are 
the intrastate carriers now operating in California and Texas at'le.ast 
as profitable as the trunk carriers as a whole, but their safety records 
compare quite well with those of the trunk carriers.^*^^ 

Finally, the supplemental carriers are subject to considerably less 
rate regulation than the trunk carriers, but on the basis of fatalities per 
billion passenger-miles, their safety records are better than those of 
the trunks for nine of the past eleven years.^^^- 

Overall, then, there is reason to challenge the need for regulation 
as practiced by the CAB over the interstate trunk airline industry. 
The law guiding the CAB is a product of times very different from to- 
day : economic conditions were depressed, and there was a widespread 
feeling that rigging prices at noncompetitive levels was conducive to 
recovery. Furthermore, the airline industry was not economically vi- 
able without subsidies. It might at the time have made some political 
sense to control entry and faros so as to minimize the need for such 
subsidies. But the situation is different now. in that the trunk airline 
industry is viable and has been off subsidies for many years. If public 

^^^ Kennedy Report, p. 73. 

^•^ See prepared statement of G. L. Seevers and J. C. Miller III, Kennedy Hearings, 
p. 65. 

^* Slmat. Hellio?en, and Eichner, vol. I, p. 2. 
'" Kennedy Report, p. 73. 

83-944—78 10 


policies ought to be changed, how should they be changed? That is the 
topic of the final section. 

Policy Alternatives 

There are several possible alternatives for dealing with the problems 
described in the previous sections. One would involve better, perhaps 
tighter regulation, with the goal of achieving a better trade-off between 
fares and service quality (a special and extreme case of this would 
be nationalization). Another alternative would be complete and im- 
mediate deregulation. Yet another alternative is partial and gradual 
regulation. Finally, there is the matter of direct subsidies for un- 
profitable routes (felt, nevertheless, to be socially desirable) under 
any one of these schemes. All alternatives relate only to regulation 
of fares and firm entry ; no consideration is given to deregulation of 
safety, noise, or pollution standards, largely because there is little or 
no economic evidence that these last types of deregulation are desirable. 


A regulator could, by setting fares at an appropriate level (i.e., 
basing them on a higher load factor standard than currently prac- 
ticed) achieve some of the results of deregulation. Or, a more drastic 
measure could be taken to achieve higher load factors: direct regu- 
latory control over the amount of service (i.e., capacity) provided on 
a given route. Let us consider each of these alternatives. 

The CAB has itself given good evidence that when (for political 
or other reasons) it wants low coach fares, it can get them : regulated 
fares to Puerto Eico and Hawaii are considerably lower than on equiv- 
alent routes within the continental United States.^^'' In addition to 
that, as we have seen, the CAB recently completed the process of at- 
tempting to eliminate the cross-subsidy from coach to first-class serv- 
ice by adjusting each type of fare appropriately. Indeed, for all the 
criticism it has incurred, the CAB cannot be said to be totally un- 
responsive to considerations of economic efficiency. However, it is per- 
haps inevitable that a regulatory agency will be subject to political 
pressures and tend to protect the interest of producers and of special 
interests such as small communities at the expense of the public in 
general, which is less likely to exert political pressure on behalf of 
lower fares for itself.^'^^ Moreover, markets and economic conditions 
change, and it seems to be more difficult for regulations to allow prices 
and services to change to meet the needs of the traveling and shipping 
public than for the market to do the job itself. That this is so is docu- 
mented not only by the experience of airline regulation, but also by 
considerable evidence from other regulated industries, some of which 
are discussed elsewhere in the present volume. 

All these arguments apply even more strongly to the second possible 
"reform" mentioned in the previous paragraph: CAB regulation of 
service quality as well as fares. Based on evidence from fare regula- 
tion, it is likely that CAB regulation of schedules would make trunk 

i«7 Ibid, pp. 49-51. 

i«8A precise economic theory describing why and how a regulator will do this has 
been developed by Sam Peltzman. See "Toward a More General Theory of Regulation," 
Journal oj Law and Economics (August 1977). 


airline markets even less responsive to the needs of the public than 
they are now. 

thus, the first-mentioned alternative of better regulation, that of 
basing fares on higher load factors and no cross-subsidy from coach 
to first-class, would be preferable to the current situation. On the other 
hand, direct regulatory control of schedules could easily make things 
worse. . . 

An even more extreme way of restricting service quality rivalry 
would be to allow the airlines to do the regulating themselves, i.e., 
grant them the right to make "capacity agreements," cutting back on 
service through accords worked out among the carriers. This arrange- 
ment, which has occasionally been condoned by the CAB, most recently 
in the 1978-74 energy crisis', is even less likely to work in the interests 
of the traveling public than control of service quality by the CAB. 
The reason is simple : the airlines would have no incentive to take into 
account the interests of the public in their capacity-restriction agree- 
ments; their only incentive would be to achieve higher profits, and 
that is likely to result in substantially worse service than free markets 
would generate.^^^ 


This form of de facto deroirulation has been pursued bv the CAB 
over the nnst three vears, with a sharp acceleration since the appoint- 
ment of Alfred E. Kahn and Elizabeth Bailey to the CAB in 1977. 

Briefly stated, what the CAB has done over the past three years is 
to take a much more permissive attitude toward fare cuts by existing 
carriers than previously. Since the appointments of Kahn and Bailey, 
it has in fact not rejected any proposed fare cut by any carrier on a 
domestic route.^^^ 

The results of these policies have been di'amatic : airlines have intro- 
duced such a plethora of discount fares that a full accounting of them 
here would be impossible. Perhaps the most widespread of discount 
domestic fares has been the Super Saver fare, available on most inter- 
state routes. Basically, this fare allows a discount of 20—10 percent, 
relative to the regular coach fare, with certain restrictions, such as the 
requirement of a round trip purchase, a minimu.m stay of seven days, 
a limited number of seats available in a given plane, and an advance 
purchase requirement of thirty days. Most of the other fare reductions 
implemented so far have also been similarly restricted with round trip 
and advance purchase requirements, and, as has been previously 
argued, are discriminatory in the sense that they do not fully reflect 
the competition observable in intrastate markets, where similarly low 
fares are available to all travellers. 

However, in the Spring of 1978, even more dramatic changes were 
occurring: two airlines. Continental and Western, were authorized to 
implement sharp fare reductions on totally unrestricted travel on main 
interstate routes. 

Continental's "Chickenfeed" fares, to become eflPective May 15, 1978, 
provide an unrestricted discount of 30 percent off regular coach fares 
during the daytime (in a special rear compartment of the plane) , plus 

169 por a more detailed critique of capacity agreements In theory and practice, see 
the Kennedy Report, pp. 112-159. 

"""Cutting Confusion in Air Fares," Business Week (April 24, 1978), p. 64. 


a 40 percent discount from regular coach fares at niglit (applying to 
all coach seats). These fares apply to all of Continental's routes.^'^^ 

Similary, Western is implementing an across-the-board reduction 
of 35 percent in coach and first class during the day, with a 48 percent 
dicount for night coach, on its Miami-Los Angeles route. These fares 
will be matched by Western's competitor on the route, National.^^^ 

In order to expedite further airline fare-change proposals, the CAB 
has now proposed a greatly-simplified set of administrative pro- 
cedures, which would allow fare decreases of up to 50 percent per year, 
and increases of up to 5 percent per year, by existing carriers on a 
given route, without going through the delays of a fare-change 
procedure. These procedural changes will be made without new 

It is worth noting that the fare reductions being implemented by 
Continental and Western are of the order of magnitude suggested 
previously in this chapter as being consistent with "deregulated" mar- 
kets. If these trends continue and the low fares spread to other markets, 
one might argue that no legislative reform is needed to achieve the full 
benefits of deregulation. 

However, there are several reasons why legislation is needed to 
assure that consumers get the full benefits of deregulation in trunk 
airlines, and that these policies continue over time. 

First, while a strong case can be made that the policies currently 
being pursued are indeed legal, they do represent a sharp change 
from previous policies, and present changes could be challenged 
in courts on the grounds that the new policies are not consistent 
with the laws as they were interpreted between 1938 and the mid- 
1970's. Although this is potentially to some degree a problem with fare 
reductions, it could be even more of a problem with regard to proposals 
for entry by new carriers. Thus, several carriers with no existing 
certified interstate trunk routes have proposed to enter a number of 
trunk routes at fares considerably lower than existing ones.^''* The 
CAB has not yet ruled on those cases, and it is not clear what decision 
will be made. However, given that the CAB has not allowed totally 
new carriers to enter trunk routes in the past, a strong precedent exists 
to deny entry to such routes. As will be argued below, the entry of 
new firms on to trunk routes could play a critical role in getting fares 
down to "competitive" levels. 

To assure that some new carriers will have an opportunity to enter 
the trunk airline market without court challenge, legislative reform 
is highly desirable. 

A second reason for legislative reform is that the make-up of the 
CAB could easily change, and appointees more sympathetic to higher 
fares could easily regain control. Without legislative reform, there is 
little guarantee against a return to the fare levels of 1974 and earlier 
once the tenures of certain existing members have expired. 

1^ Ibid., p. 64. 

i"^2 Ibid., p. 64. 

i'" "Airline Regulation Has a Long Way to Go," Business Week (May 8, 1978), p. 4.3. 

1'^* One of the most important of these pending eases is that of Midway Airlines and 
Southwest Airlines which wish to provide short-to-medium-haul service between Chicago 
and various midwestern cities, at fares 12-64 percent below existing regular coach fares. 
See U.S. Senate, Committee on Commerce, Science, and Transportation, Report on Amevd- 
ing the Federal Aviation Act of 1958 (Washington, U.S. Government Printing Office, 
February 6, 1978), pp. 103-105. 


In short, there are some very good reasons why legislative reform is 
needed to buttress and expand the policies already being pursued by 
the CAB. Alternatives for legislative reform will be considered below. 


^\niat has been said of "better" regulation is likely to apply all the 
more to nationalization. Here again, political considerations are likely 
to outweigh the needs of the market in the operation of the enterprise, 
and it is not likely to be as responsive to the needs of shippers and 
travelers as competing private firms. Furthermore, there is no evi- 
dence that the nationalized airlines in other countries are in any way 
superior to the privately-owned, regulated ones in this country in terms 
of efficiency or in terms of the fares they charge. To the contrary, evi- 
dence collected by M. Straszheim would indicate that foreign nation- 
alized carriers have been historically more costly to operate than pri- 
vately-owned American ones. Thus, in the late 1950's, the operating 
deficits of many Western European carriers accounted for a large part 
of national output, ranging from 0.6 per cent in Great Britain to an 
astounding 3.0 per cent in the Xetherlands.^^^ 


On the basis of the arguments made in previous sections, there is 
much to be said for complete and immediate deregulation of trunk air- 
lines : there is little about the industry which would seem to require the 
regulation of fares and firm entry or route authority, and the benefits 
would seem to be great from the viewpoint of the public. 

There are, however, several major disadvantages to complete and 
immediate deregulation. The first and most important relates to issues 
of equity. It was stated in Section II that while airline investors, sup- 
pliers, and workers should, overall, be expected to benefit from deregu- 
lation, nevertheless there are likely to be some short-run losers, mostly 
specifically less efficient firms, or firms on routes with demand which is 
unresponsive to price. If deregulation is immediate, the changes could 
be painful or disruptive, and the political feasibility of deregulation 
would be jeopardized. Second, to the extent that politically undesirable 
consequences of regulation occur (such as loss of service to some small 
communities) it is appropriate that the change occur in an "orderly" 
manner (i.e.. that government authorities should have time to allow 
for subsidized service where it is needed) . To achieve "orderly" change, 
gradual deregulation is preferred. Third, as pointed out by the Ken- 
nedy Committee report, managers in the industry must have time to 
adapt to conditions of deregulation.^'^ For this reason, it is not clear 
that the benefits would come immediately even if deregulation were 
complete and immediate. It takes time for now firms to enter markets, 
or for existing ones to enter new markets : planes must often be built 
or reconfigured: route plans must be made, and profit calculations 
done. In this context, it is worth noting that the full benefits of unregu- 
lated markets in California, most specifically from PSA, did not come 

i~ Spo ^Mfih^on Sfraszhoim. The International Airline Industry (Washington, The Brook- 
in?s Institution. 1970). p. 27. 
^"« Kennedy Report, p. 74. 


until 1959, over ten years after it first entered the market for air trans- 
portation (when it instituted service with modern aircraft). As a 
result, it is not clear that immediate deregulation has too much of an 
advantage over gradual deregulation in terms of when the benefits 
would make themselves felt, while gradual deregulation has several 
advantages over immediate deregulation as enumerated above. 


This is the sort of deregulation which has the most appeal from sev- 
eral viewpoints, and it is the sort of deregulation called for by the 
most popular of the current legislative proposals.^" The markets in 
California and Texas are not now and never have been totally dere^- 
lated. Even in California before 1965 the California Public Utilities 
Commission had some control over air fares. Our experience with intra- 
state airlines has all been in an environment of partial deregulation. 

Many combinations of partial or gradual deregulation policies could 
be instituted. We consider only the alternatives that have been seriously 
proposed, or those which have particular appeal in terms of their 
ability to achieve efficient results without adverse consequences. 

Price Deregulation without Entry Deregulation 

One possibility for partial deregulation is to deregulate rates with- 
out deregulating firm entry on a given route. This is likely to have 
serious adverse effects in the form of higher fares on some routes: 
without entry (or threat of it), existing carriers would probably be 
able to coordinate their rates at high levels, for most routes are highly 
concentrated, making coordination all too easy. This form of "dereg- 
ulation" could easily produce market performance worse than the 
existing setup. 

Deregulation of tSelected Routes 

Another possibility is deregulation of rates and of firm entry on a 
few routes on an experimental basis, with expansion to other routes 
if the experiment is a success. This approach is flawed on several 
counts. First, there is the possibility that whatever route or routes 
may be selected might not be representative of all routes, so the results 
of the experiment might be misleading. Besides, we already have the 
results of some "experiments" in the form of the intrastate routes. 
Another problem is that the deregulated route could be an object 
of capacity-dumping or predatory pricing on the part of airlines on 
regulated routes. It is not clear what evidence there is, theoretical 
or empirical, to support this possibility : predatory pricing and simi- 
lar behavior are not generally rational, and there is little evidence 
of it frequently occuring in the unregulated sectors of the U.S. 

1" Numerous bills for deregulation were introduced in the 94th Congress, but the proc- 
ess of compromise has sharply narrowed those introduced in the 95th. More specifically, 
in April 1978, the Senate passed the Cannon-Kennedy Bill, largely a compromise between 
separate bills introduced by each Senator in the 94th Congress. The provisions of this 
Bill will be discussed in some detail in this section. As the last revision of this chapter 
was being written. The House Aviation Subcommittee was still workin? on a similar bill 
proposing airline deregulation. Because none of the details of what was to come out of 
the House were available as of this writing, I shall not comment on it. Rather, emnhasis 
will be placed on the components of the Cannon-Kennedy Bill which should definitely 
be carried over into the House bill. 

178 p. M. Scherer (op cit., pp. 27.S-278) surveys all the major cases of alleged predatory 
pricing up to the year 1968, and finds no persuasive evidence that it actually occurred. See 
also Roland Roller. "The Myth of Predatory Pricing: An Empirical Study," Antitrust 
Law Review (Summer,1971). 


Furthermore, there is little evidence that the certificated carriers 
participating in the California and Texas markets have engaged in 
extensive capacity-dumping or predatory pricing against the intrastate 
carriers in those routes.^"^ There are, however, other potential ob- 
jections to this form of partial or gradual deregulation. By tliis route, 
deregulation in all markets might be delayed interminably by hearings 
to determine if experiments were "successful'' and by appeals of what- 
ever results the hearings achieved. This could slow the process of de- 
regulation to a standstill. 

Deregulation of High-Density Routes 

Yet another possibility^ for partial deregulation would be to deregu- 
late only those routes (probably high-density ones) where low con- 
centration and competition would appear most likely. This approach 
has much to say for it. High-density routes are inded those which can 
"support" the most firms, and perhaps the most successful of the intra- 
state routes have been high-density ones. However, there is evidence 
that deregulating these routes alone could deprive many passengers 
of the full benefits of deregulation. There is, in fact, substantial evi- 
dence that unregulated medium- and low-density routes can support 
excellent service at fares no higher than those now charged. This 
evidence comes from some intrastate routes, such as the Houston- 
Harlingen route in Texas and the Fresno-Los Angeles and Fresno- 
San Francisco routes in California, and from the many commuter 
routes vrhich are totally unregulated by the CAB. How does this 
come about ? In a number of cases the carriers serving these loosely- 
regulated low-density routes have a monopoly or near monopoly on 
the routes served. There are two basic reasons for this good market 
performance in the face of an apparent monopoly. 

The first stems from the fact that if prices exceed costs b}- too much 
(or if service levels deviate from optimal ones too much), new fi.rms 
will enter. The second reason is that most lower-density routes tend 
to be relatively short in haul,^^'' and as a result there is substantial 
surface competition (buses and especially private autos on low-density 
routes). If the fare from Fresno to Los Angeles is too high, most 
travelers, business or pleasure, will not hesitate to drive. As a result 
of this surface competition, most low-density routes which appear to 
be monopolies are not monopolies at all. All this evidence suggests 
that there is no reason to deny those traveling on low- to medium- 
density trunk routes the benefits of deregulation. (We are talking here 
only about those routes which are, in fact, economically viable. The 
nroblem of subsidizing service on unprofitable routes^ is discussed 

Gradual Deregulation of All Routes 

A final possible form of gradual and partial deregulation is to 
allow fares to change and firms to enter in all markets, but to expand 
their abilities to do so only gradually. This is, in fact, the approach 

i'^ It has. of course, been the Intrastate carriers who were the main price cutters in 
these markets, and thoy had no outside capacity to dump, or profits from elsewhere to 
cross-<!ubsidize predatory prices in the intrastate markets. 

1*^ The reason for this stems from the logical route structure of the T^.S. and most other 
nations. For example, instead of flying on a nonstop flight from Bangor, Maine to Mon- 
terey. Calif., a passenger will, in all likelihood, take a short-haul low-densitv flight from 
Bangor to Boston, a long-haul, higher-density flight from Boston to San Francisco, and 
another short-haul low-density flight on to Monterey. 


taken by more or less all recent proposals for regulatory reform in 
the airline industry.^^^ 

The idea behind this is to allow fare change and firm entry to occur 
gradually, so as to minimize financial and service "disruptions" during 
a "transition period" between the existing system and a system con- 
ferring all the benefits of the intrastate systems mentioned above. The 
full benefits of deregulation, however, would occur only after a period 
of some years (say, four to ten years, by most current proposals). As 
previously stated, freedom to achieve both firm entry into a market, 
and fare change are important from the viewpoint of the goals devel- 
oped in this chapter, and it is appropriate to examine in some detail 
the alternatives in the case of each. 

Firm Entry 

Currently an airline may not serve a route without a certificate from 
the CAB indicating that public convenience and necessity requires its 
services on that route. To achieve entry onto a new" route, a firm, 
whether it be a carrier already certificated for scheduled service on 
other routes, or a firm not currently providing scheduled service must 
prove to the CAB that public convenience and necessity require its 
addition to that route. As previously stated, the CAB has not allowed 
the entry of any entirely new truck carrier since the 1938 Act was 
passed, although it has granted numerous route expansions by existing 

To assure that after fare deregulation, entry on to various routes is 
sufficient to get fares down to a more "competitive" level, there are 
several steps which can be taken. 

One possible step involves a change in the Statement of Purpose 
(discussed earlier in this chapter) to require that in pursuit of its goal 
of "efficiency, innovation, and low prices" in air transportation, the 
Board should rely the forces of "actual and potential competition." 
The Cannon-Kennedy Bill in fact does this.^^^ This change is a most 
important one, for it would give the CAB a clear legislative mandate 
to pursue more permissive policies as regards fare flexibility and firm 
entry. The desirability of such a mandate has been discussed above. 
But a change in the Statement of Purpose alone is not likely to be ade- 
quate over the long term. There is no guarantee that future members 
of the CAB will be as well disposed toward fare flexibility and route 
entry as the current ones. Furthermore, their attempts to implement a 
more liberal entry and pricing policy would be made much easier with 
more specific entry provisions. 

For these reasons, if the nation is to be assured of getting the full 
benefits of deregulation, it is necessary that reform law should clearly 
spell out and guarantee the rights of firms to enter new markets. In 
this connection, however, some questions arise. Wliat firms should be 
allov\'ed to enter what markets, at what times? If deregulation is 
indeed to be gradual but eventually total, answers to these questions 
must be spelled out. 

isi See footnote 177, above. 

issThe above quotes are from the text of S. 2493, See Hearings on Amending the 
Federal Aviation Act of 1958, op. cit , p. 6. 


First, let us consider the issue as to what firms should be allowed 
to expand into what markets. Here, it is important to make a distinc- 
tion betA^een two types of entry, mentioned above, but not discussed 
in detail : entry of existing, certificated firms on to new routes, and 
entry of firms currently without scheduled certification into scheduled 

Does it make any difference whether botli types of entry are allowed ? 
Xeither economic theory nor empirical evidence allow a complete an- 
swer to this question, but it is possible to make some reasoned guesses. 
We start by considering two possible viewpoints on tlie issue, first that 
it is not necessary to allow carriers without current scheduled certifi- 
cates into deregulated markets, and second that it is necessary. 

A holder of the first viewpoint would argue that it makes no differ- 
ence whether entry on a given route is by a carrier with certification 
elsewhere or a new one. The argument here is that currently in this 
country there are ten domestic trunk carriers, plus nine local service 
carriers, and that if these carriers are allowed to expand gradually on 
to new routes, that will be adequate to achieve the fruits of competi- 
tion. Thus, if political considerations prevent reform from allowing 
rapid and (effective entry of carriers witliout current interstate certifi- 
cation (such as supplementals and intrastate carriers), that is no prob- 
lem, because existing certificated carriers should provide adequate 
price competition to achieve efficient markets in both the price and 
service dimensions. Evidence to support this comes from the fact that 
when given free reign in service competition, plus the limited entry 
freedom which the CAB has allowed, the carriers liavc been unable 
to coordinate frequency or capacity competition so as to earn excess 
profits, so why should tliey be able to do any better with fares ? 

More recent evidence supporting the viewpoint that entry of existing 
trunk carriers to new routes is an adequate spur to competition comes 
from the behavior of the trunk carriers under the less restrictive fare 
policies of the Board over tlie past year. As previously stated, numer- 
ous fare reductions have been occurring, and the amounts are consider- 
able, ranging up to 45 percent. 

A holder of the second viewpoint might question whether competi- 
tion among the existing certified carriers will be adequate to assure 
the full benefits of deregulation. More specifically, although the trunk 
carriers have indeed reduced some fares sharply over the past year, 
the vast majority of these reductions liaA'e been highly discriminatory 
in nature, with advance-purchase requirements and round-trip restric- 
tions. Even in cases where fare reductions have been available to all 
travellers (i.e., rontinontal's Chickenfeed Fares), tickets sold are 
restricted to well-below half available seats on flights (the only excep- 
tion to this so far is Western Airlines' California -Florida fares, previ- 
ously mentioned). 

The most important evidence that trunk air fare cuts have so far 
been noncompetitively discriminatory comes from the intrastate mar- 
kets: althoujrh these mnrkets do indeed hnve peak and off-peak fares 
(quite justifiable theoi-etically based on competitive markets) they 
have no round-trip restrictions or advance purchase requirements. And 
low fares are not restricted to some small fraction of the plane cabin — 
they are available to all passengers. These non-discriminatory fare 


cuts, incidentally, are also being proposed by those firms currently 
petitioning to enter anew into interstate markets.^^^ 

The point of this evidence on intrastate markets is that oftentimes 
it is only firms totally new to trunk markets which have shown a will- 
ingness "to cut fares on an unrestricted, nondiscriminatory basis. For 
that reason, it can be argued that the entry of firms totally new to 
trunk air transportation, or at least the threat of such entry, may be 
critically important to the achievement of nondiscriminatory low 

There are more theoretical argimients, as well, why carriers without 
current scheduled certification should be allowed to enter scheduled 
service on interstate routes. The trunk carriers could, for a time at 
least, even after more or less complete deregulation of fares, agree 
tacitly not to "poach" on each other's territory, recognizing the inter- 
dependence of their decision. Furthermore, as F. M. Scherer and 
other students of unregulated oligopoly in the U.S. have argued, prices 
always tend to be much easier to coordinate than product quality: 
price changes can be matched quickly, whereas product quality changes 
cannot; price changes are easily noticeable and quantifiable, whereas 
product quality changes are not.^®* For these and other reasons, the 
airlines may have an easier time coordinating fares in a deregulated 
environment than they have had coordinating service quality. 

No one can claim to have "hard" evidence to determine which of 
these views is right. Nevertheless, it can be argued that on the basis 
of the existing trunk and local service, new carriers should be allowed 
entry on to interstate routes. If, for political reasons, it is desired not 
to allow "wide open" entry of anyone who wants in (subject to safety 
provisions), another approach would be to allow only existing supple- 
mental and intrastate carriers into scheduled service. 

Given that there are now eight supplemental carriers, plus four 
major intrastate carriers, this group provides a substantial pool of 
potential entrants on to scheduled trunk routes.^^^ 

Once it is established that there should be entry guaranteed on trunk 
routes, for both existing trunk carriers and for "new" firms, the matter 
arises as to just how open and free such entry should be. From the 
view of the economic analysis presented above, it is difficult to justify 
any restrictions (other than safety ones) for trunk routes. However, 
many existing carriers have deep fears of the potential consequences 
of totally open entry, even by a restricted pool of potential entrants. 
Therefore, all legislation introduced in the 94th and 95th Congresses 
has had provisions for limiting automatic entry, especially during the 
first year or two after regulatory reform has been implemented. 

The Cannon-Kennedy Bill, passed by the Senate, has especiall}^ 
complex safeguards to prevent overentry or excessive competition 

1^3 More specifically, World Airways, a charter carrier, has on several oceassions at- 
tempted to enter transcontinental markets at unrestricted fares well below existing ones 
(Keeler, 1972. p. 416). Furthermore, two airlines. Southwest, the Texas Intrastate car- 
rier, and Midway Airlines, a newly-formed carrier, have petitioned to enter sixteen dif- 
ferent markets between Chicago and various Midwestern cities. See "Flying Cheap . . . 
and Full," Forhes 118 (December 1, 1976), pp. 45-46. 

i«4 Scherer, pp. .S44-.S47. 

18.-! The Cannon-Kennedy Bill provides for some automatic entry by supplementals and 
intrastates. See Report on Amending the Federal Aviation Act of 1958, op. cit., p. 13. 
Furthermore, it allows for new charter carriers to be certified without demonstration 
that public convenipnce and necessity require it, after 1982. This should allow an expanded 
pool of supplemental carriers. 


stemming from overentry, safeguards which merit consideration in 
some detail. ^^^ 

First, the number of new routes to which firms would receive auto- 
matic entry in a gi^-en year is restricted. During 1979 or 1980, each 
designated carrier (trunk, local service, intrastate or supplemental) 
could receive automatic entr\' on to only one nonstop route segment, 
not to exceed 3.000 miles. During 1981. and for each year thereafter. 
each such carrier would be free to enter two such routes, totalling no 
more than 3.000 miles between them. 

Second, these automatic entry provisions are further restricted by 
the ability of carriers to prevent the entry of other carriers on a limited 
number of existing routes. More specifically, in 1979, 1980, and 1981, 
each carrier would be able to specify three of its routes from which 
automatic entry should l)e barred. However, these restrictions are de- 
signed to be transitional, dwindling to two segments per carrier in 
1982. one in 1983. and none at all thereafter. 

Third, the Bill protects local service carriers and small trunks from 
even this mucli competition from automatic entry during the 1979-81 
period. Specifically, it states that no local service carrier shall have 
more thpn two of its route segments open to such entry during 1979-81, 
and no carrier providing fewer than 2o billion available seat-miles per 
year shall have more than three such route segments in its system 
cei-tified for automatic entry by another carrier during 1979, and no 
more than four such routes in 1980. Plowever. after 1981, these pro- 
visions would be eliminated. 

Fourth, the Bill provides that in 1983, the CAB shall report in de- 
tail to Congress on the effects of these entry policies, with the possi- 
bility of further legislative revisions if the existing laws have 
somehow worked unsatisfactorily. 

Finally, the Bill provides that if these provisions should cause sub- 
stantial pul)lic harm to the nation's air transportation system, or if 
it should cause a substantial reduction in service to small and medium 
cities, the CAB shall have the power to revoke or modifs' these rules 
to the extent neces.sary to rectifpy these problems. 

These are restrictive provisions on the basis of analysis previously 
presented here, they would seem to be much more restrictive than 
needed to achieve workable com))etition. But the question here is not 
whether they are more restrictive than necessary: if theso restrictions 
make the bill politically acceptable to airlines and lecrislators from 
small communities, so much the better, provided that they are not so 
restrictive as to prevent travellers and shippers from receiving the full 
benefits of competition, at least after a transition period of five vears 
or so. 

Are the entry provisions of \h^ Cannon-Kennedv bill sufTicientlv 
permissive to achieve the benefits of dereirulation arirued to exist i'n 
this cha]:.ter? It is impossible to say for sure, but there is good reason 
to believe that they are quite adequate, for several reasons. 

Fii^t. althouofh the total number of new route awards during any 
one year could be limited, especiallv at the becrinning, they should add 
up over time. After all. the total number of firms having the right to 
one or two new routes per year is over twentv. meanin^r that the'^num- 
bers of new routes entered per year could be over forty after 1980. 

i"* Ibid., pp. l.S-17. 57-68. 


Second, even ^Yithout actual new entry occurring, these provisions 
allow for the constant threat of new entry on a route, if fares are high 
relative to what competitive market conditions would dictate. The au- 
thors of the Senate Report supporting this bill were right in empha- 
sizing that these considerations of potential entry are very important, 
for if an airline sets fares too high relative to costs, it cannot avoid 
the cha-nce of new entry induced by these high fares.^^^ 

Third, it must be remembered that all these provisions relate to 
automatic entry. There is nothing in the Bill which prevents the CAB 
from alloAving further entry, if it feels that such entry is appropriate 
from the viewpoint of the shipper and travelling pul3lic. Indeed, the 
changes in the statement of purpose are designed to encourage the 
CAB to authorize such entry. 

To the extent that entry of new firms is deregulated, controls on 
fares should be eased, allowing the forces of the market to control 
them. In the case of the Cannon-Kennedy Bill, however, entry would 
not generally be totally free, and with the restrictions which it pro- 
poses, many routes could be protected from new entry for some time. 
As a result, it is important that some maximum-fare regulation be 
continued, to prevent exploitation of temporary monopolies allowed 
by this regulation. Furthermore, to allay the fears of the airlines that 
deregulation could result in destructive competition or predatory pric- 
ing, some floor on fares might make sense politically, as well, although 
it is difficult to justify it on the basis of economic evidence presented 

The Cannon-Kenned}^ Bill solves these two problems by allowing 
free movement of rates inside a "zone of reasonableness" built around 
the existing regular fare levels.^^® More specifically, it proposes that 
the airlines should not need CAB approval to lower their fares by as 
much as 35 percent or raise their fares as much as 5 percent relative 
to the level of regular fares in each class of service prevailing July 1, 
1977, or adjusted for inflation in costs using current CAB procedures 
for years thereafter. This, then, represents a "zone of reasonableness" 
around the fares as established by the DPFI, discussed earlier. To 
prevent unreasonable exploitation of this zone on monopoly routes 
and predatory behavior on competitive routes, this zone of reasonable- 
ness is qualified by two important exceptions. First the 5 percent upper 
zone of flexibility is not allowed on routes in which the carrier charg- 
inof the fare has 70 percent or more of the market. Second, fares falling 
within this boundary on the low side may be ruled "predatory" and 
prevented if it can be shown that they are not only below costs, but 
also sufficiently low to put another firm out of business. 

Of course, the zone of reasonableness relates only to those changes 
which airlines can make without CAB approval. With CAB approval, 
changes outside this zone of reasonableness would be feasible. The 
CAB is, however, directed to take into account additional principles 
in approving fares outside the zone of reasonableness. First, the Board 
is encouraged to approve fares which fluctuate based on supply and 
demand considerations, such as peakload prices.^^^ Furthermore, it is 

IS- Ibid., p. 50. 

i^'S Report on Amending the Federal Aviation Act of 195S, pp. 106-109. 

1S0 Unci., p. 108. 


directed against allowini^ fare changes which are unduly discrimina- 
tory, preferential, or prejudicial.^^° 

In combination with the entry provisions described above, how well 
will these fare flexibility provisions work to achieve the desired com- 
petitive results? 

It can be argued that they will work reasonably well. To rret some 
idea as to the extent to which a 35 percent fare reduction might achieve 
more competitive fares, consider the "hypothetical competitive fares'' 
in Table 4. Although these fares relate to 1974. they are based on the 
same DPFI principles as regular CAB fares are derived from cur- 
rently, and, because of the elimination of the cross-subsidy from coach 
to first class, coach markups from "optimal" to actual fare^ should 
if anything be lower today than they were in 1974. In any event, note 
that the markups from actual to calculated optimal fare in that table 
rnnged from 30 to 56 percent. These mean that gettinor from existing 
to optimal fares would entail reductions of 24 to 36 percent (i.e.. 
1/1.56 = .64, and l-.64 = .36). Thus, to the extent that Table 4 is sug- 
gestive of how much fares "ought" to go down with deregulation, it 
would suggest that a reduction of 35 percent or less should be adequate 
in most cases. This means that the floor 35 percent below existing 
DPFI-type regular fares should be adequately low to allow for most 
of the potential benefits of deregulation, jyrovidpd thiat the 35 percent 
reduction ultimately applies to unrestricted fares, rather than dis- 
criminatory excursion fares. (And nothing in these provisions pre- 
vents the reductions from applying to unrestricted fares.) 

On the other hand, the upward component of the zone, 5 percent 
above "regular" fares, should be adequate to allow the airlines to com- 
pensate for inflation without waiting for Board approval. (It will be 
noted here that if in fact competitive markets reduce fares well below 
the standard level, the zone of reasonableness will afford those markets 
far more leeway to adjust quickly to inflation than 5 percent). 

Overall, then, while a zone of automatic downward flexibility of 
greater than 35 percent would be desirable, "^ would appear that the 
fare provisions of the Cannon-Kennedy Bill should be adequate to 
permit the market to achieve fares of a reasonably competitive level 
on most routes. These fare rules, combined with the automatic entry 
rules discussed above, allow for a gradual transition of the trunk air- 
line industry to the same workably competitive state enjoyed by the 
various intrastate markets. There are, however, a number of possible 
side-effects and other considerations relating to this move to greater 
competition, and they are the next topics for consideration. 

Other Issues Conupcted with Deregulation 

In addition to --eform of rate and entry regulation, there are some 
other peripheral issues which are worth discussing here, because they 
are closely connected with deregulation of fares and entry. 

Service to S^niall Communities 

It was argued earlier that CAB regulation has brought about little 
cross-subsidization from high- to low-density routes. Furthermore, 
the use of cross-subsidization as a regulatory practice is questionable 
on grounds not only of economic efficiency, but also of equity. As 

iw Ibid., p. 109. 


Caves stated it so aptly some years back, "there is iio reason why 
impoverished grandmothers fly in or from New York to T.»os Angeles 
should be the ones to subsidize well-off businessmen traveling between 
small towns.'* *®^ Thus, it is not unreasonable that a regulatory bill 
sliould make provisions for a carrier to drop service which it finds un- 
profit<able. There is evidence that that has been done already, for 
the most part, as pi*eviously stated. But there should be no room for 
even a vestige of cross-subsidization in deregulated rates. 

As a result, a carrier should be free to discontinue unprofitable serv- 
ice. If, on the other hand, a carrier is required by CAB policy to 
continue sen-ice which it deems unprofitable and wishes to discon- 
tinue, it is appropriate that the carrier be compensated for that service, 
in the manner in which local service can*iers are currently compen- 
sated. The Cannon-Kennedy bill specifically provides for this. Fur- 
thermore, it allow complete abandonment of a route by a certified 
carrier, provided another carrier can be found to provide the ser\-ice 

Given that comnuiter carriei*s have managed to provide breakeven 
service on routes lower in density than the subsidized local service 
carirers service, it is likely that subsidy requirements to comnnmi- 
ties which would otherwise lose service after deregulation will be low 
or nonexistent (excluding the special case of very low-density routes 
in Alaska), provided that a commuter carrier gives the service."^ 

Service to such communities could be guaranteed by reform legisla- 
tion, at least for a few years, so as to make deregulation more politi- 
cally palatable for small communities. The Cannon-Kennedy Bill pro- 
vides for such a guarantee, supported by Federal funds, for a period 
of 10 years after the enactment of the reform legislation. 

Also in connection with the commuter carriers, it will be recalled 
that they currently face a 30-passenger or 7,500-pound (net) capacity 
constraint if they are to avoid CAB regulation. To allow for better 
and more economical commuter service, where it is possible, the Can- 
non-Kennedy bill proposes an increase in this maximum size to 56 
passengers or a gross takeoff weight of 40,000 pounds. This change w^ill 
make commuter service more efficient and attractive to communities 
which do not receive trunk service.^®* 

Antitrust Exemptions — Trade Agreements 

Current law allows the CAB to grant airlines certain rights to 
enter into agreements which would otherwise involve antitrust vio- 
lations. These agreements involve such things as capacity restriction, 
as well as fare and schedule coordination. The CAB has generally used 
these powers sparingly.^^^ Xeveilheless, such practices could under- 
mine the functioning of the competitive markets advocated here. For 
example, if airlines were allowed to set up capacity agreements, they 
could, by controlling capacity, keep fares artifically high without nec- 
essarily colluding explicitly on fares. 

It is, therefore, appropriate to subject the airlines to more or less 
the same antitrust laws regarding collusion in restraint of trade whicli 

"leaves, p. 411. 

^^- tfcfjort on Amending The Federal Aviation Act of 1958 ,pp. SS-94. 
"^ See the discussion of the amoiint of cross-subsidy, above, in this section. 
iM See Repo't on Amending the Federal Aviation Act of 1958, p. 25. 
"* One recent time It has allowed them Is 1974, when, because of the energy crisis, 
it allowed the airlines to work out flight cutback agreements. 


the unre^lated sector of the economy faces, unless there is some dire 
reason to do otherwise (i.e., a national emergency requiring quick over- 
all cutbacks in civilian air transport). The Cannon-Kennedy Bill does 
this for price-fixing and capacity-restriction agreements.^^^ 


The current law exempts air carriers from Federal antitrust merger 
laws, and places the jurisdiction of such mergers with the CAB. The 
CAB is already under legal guidelines somewhat similar to, but con- 
siderably less restrictive than, those of the antitrust laws, in that the 
Board is directed to approve proposed mergers unless they are "not . . . 
consistent with the public interest." But the law also provides "that 
the Board shall not approve any . . . merger . . . which would result 
in creating a monopoly or monopolies and thereby restrain competi- 
tion or jeopardize another carrier not a part of the . . . merger." ^^^ 
The Board is also directed to give some consideration to Section 7 
of the Clayton Antitrust Act, which forbids mergers "where the effect 
may be to lessen competition." ^^^ But it can be argued that while some 
of the CAB's guidelines regarding mergers are similar to antitrust 
provisions, they are somewhat more permissive than the antitrust 
laws.^®^ If, in fact, the forces of competition are to play a much more 
important role in achieving efficient airline rates, then the merger pol- 
icy should be inclined further toward promotion of competition, as it 
is in the unregulated sector of the economy. 

Thus, reform of airline merger laws, to make them basically the 
same as those faced by the rest of the unregulated sector, should be an 
integrated part of any deregulation law. Still, the airline industry has 
some unique characteristics; among them is the possibility of geo- 
graphic interconnections which might allow the benefits of service 
improvements from mergers to outweigh the costs in lessened compe- 
tition. As a result, it might be desired to allow an additional antitrust 
provision in the case of the airlines, one which allows mergers when 
the lessening of competition is outweighed by potential service 

Dispatch of Regulatory Proceedings 

In the past, the CAB has sometimes made regulatory policy simply 
by delaying action and ultimately ignoring petitions. The most notori- 
ous case of this is that of World Airways in 1967, ^^nhen it proposed to 
enter a transcontinental market with a fare of $79.50, onlv 55 per cent 
of the CAB-regulated face.^oi The CAB's reaction was to ignore this 
petition for three years and then dismiss it as stale. Under all pro- 
posed legislation, the CAB will continue to play at least a token role 
in petition of route entry on the part of firms. If the CAB were 
allowed to stall as it did in 1967, it could effectively sabotage any 
such deregulation. One possible solution, of course, would be to totally 
eliminate any regulatory test of fitness, willingness, and ability of a 
carrier to enter a market. But this may be impossible or inappropriate, 
especially during a transition period. As a result, regulatory reform 

i9« See Report on Amending the Federal Aviation Act of 1958, d. 82. 
"7 Douglas and Miller, p. 201. 
-es Coleman, p. 60. 
i» Ibid., p. 60. 

200 The Cannon-Kennedy Bill basically does this, although it keeps Jurisdiction for alr- 
fofo^^^^^^o*^^^^^ under the CAB. See Report on Amending the Federal Aviation Act of 
1958, pp. 78—80. 

201 See footnote 175, above 


should place a strict time limit on CAB rate and entry proceedings 
such that if the CAB does not rule within a specific amount of time, 
the petition is automatically granted. There is something to be said 
for a shorter deadline, such as six months from receipt of a written 
petition. Delays on entry cases could be costly not only to consumers, 
but also to producers, in that they impose uncertainties which make 
planning difficult for all carriers involved. The Cannon-Kennedy Bill 
provides for a six-month deadline of the sort mentioned above. 

Employee Protection 

A significant provision of the Cannon-Kennedy Bill is a program 
to assure that certain relatively senior airline employees will receive 
special government compensation if they are laid off because of seri- 
ous financial problems brought on their employers by deregulation.^^^ 

Before considering this provision in more detail, it is worth con- 
sidering the appropriateness of such a measure in principle. As previ- 
ously stated in this chapter, there is good reason to believe that if 
anything, deregulation of the trunk airlines could bring about an 
increase in the demand for labor and capital in that industry. Further- 
more, given that civil aviation is and has been for some times a 
rapidly-growing industry one would expect that normal industry 
growth could over a relatively short time compensate for tlie effects 
of any temporary, one-shot contraction which might conceivably be 
induced by deregulation. For these reasons, one might question the 
desirability of a job protection provision such as the one proposed. 
Nevertheless, if the program is unlikely to be an expensive one, there 
is much to be said for having it. Most especially, if such a provision 
would allay fears of airline workers about deregulation, thereby mak- 
ing deregulation more politically attractive to organized labor, that 
is good reason to include it. 

The employee protection program proposed in the Cannon-Kennedy 
Bill is not an extravagant one, and it is unlikely that it could be 
expensive. Basically, it provides that if a carrier suffers a loss of 15 
percent or more of its traffic in a given year, within 10 years, of 
enactment of the regulatory reform legislation, and if this reform 
appears to be the cause, then certain employees of that carrier will be 
entitled to government compensation. To qualify for such assistance, 
employees would have to have worked for the relevant airline for at 
least 4 years, and compensation could last for no more than 3 years. 
It is also stipulated that total compensation to a given worker should 
be less per period of time than the wage previously received in the 
position from which the employee was laid off. 

Given that large-scale layoffs of the sort dealt with by this program 
are unlikely to occur, the cost of the program is likely to be negligible. 
However, according to estimates prepared by the Congressional 
Budjiet Office, the cost of the program is unlikely to be large even 
under the worst of scenarios.^^^ It estim.ated that with a 20 percent 
contraction in work force for a large trunk airline, the total govern- 
ment liability would be about $30 million. For a small trunk, the 
equivalent number would be $9 million, nnd for a typical local service 
carrier, the liability would be only $3 million. Since a 20 percent 

202 See Report on Amending the Federal Aviation Act of 1958, pp. 46-49, 113-117. 

203 Ibid., p. 122. 


contraction would be a most severe one, and since even under the 
worst of circumstances we should not expect many airlines to be sub- 
ject to such a contraction, it follows that this job protection program 
should not be expensive, especially in comparison with the potential 
benefits of deregulation already discussed. 

O verall Assessment of Proposed Legislative Reforin 

In sum, the Cannon-Kennedy Bill covers very well all the needed 
reforms for interstate airline regulation. On the basis of the analysis 
and evidence presented here, it should be enacted as law. At the time 
the last version of this chapter was being written, the House Aviation 
Subcommittee was working on a similar deregulation bill. Although 
no details on it were available, reports indicated that the House bill 
coming would have more restrictive provisions for route entry than 
the Cannon-Kennedy Bill.^°^ Although it is impossible to be certain 
just what legislative provisions are needed to achieve the full benefits 
of deregulation, the evidence presented here suggest strongly that the 
entry and fare flexibility provisions of the Cannon-Kennedy Bill 
should be preserved. It is only g-uaranteed with the entry and rate flex- 
ibility, provided for at adequate levels in the Cannon-Kennedy Bill, 
that we can be sure that travellers and shippers will achieve the full 
benefits of deregulation. Similarly, any House version should also in- 
clude modification of the Statement of Purpose, including the goal of 
low prices for travellers and shippers, to be achieved through the 
forces of market competition. Such a change could be important in 
assuring that the CAB's current drive for more competition and lower 
fares will not be stopped in the courts. 

Any House bill which does not contain at minimum these provisions 
regarding fare and entry flexibility, plus changes in the Statement of 
Purpose, will run a high risk of failing to achieve the benefits of de- 
regulation which this chapter has argued are so important. 

CoxcLUDiXG Comments 

Economists do not often agree among themselves on policy issues. 
Indeed, the present chapter has pointed out a number of disagreements 
about the effects of current airline regulation. Where such disagi-ee- 
ments have existed, it has been the aim of this study to reconcile them 
and arrive at a reasoned conclusion. But on one most important issue, 
there are no disagreements to reconcile: many academic economists 
with widely differin^y political views and research methodologies have 
analyzed the economic efl'ects of CAB regulation over the past fifteen 
years, and they have all come to a common basic conclusion — that the 
trunk airline industry would function far more efficiently (with lower 
fares) than it currently does with less regulation and more competi- 
tion. Seldom has the economics profession achieved the unanimity on 
an issue that it has on this one. 

Government organizations, both the executive and legislative 
branches, and indeed, in the Civil Aeronautics Board itself, have con- 
ducted their own studies of airline regulation, and they have ariived 
at by and large the same conclusions. 

-. ^^ '^*'"*^ ^* ^"r, "Scare Talk about Deregulation," Wall Street Journal 98 (May 9, 
1978), p. 20. 

83-944—78 11 


The Senate has now passed a bill Avhich offers much opportunity 
for the nation to benefit from deregulated air transport, while safe- 
guarding it from the major hazards. 

The time has come for the House to pass a similar bill, and for the 
President to sign these reforms into law. 


Calculations of Welfare Effects of CAB Regulation 

This appendix deals with several closely-related topics discussed 
in the text, all concerned with analyzing and to some extent measuring 
the welfare impact of CAB regulation of domestic trunk airlines. 

The first section deals with the issue as to whether the lower fares 
likely if deregulation occurs will tend to increase or decrease flight fre- 
quency and overall capacity provided on a given route. An important 
part of this section is concerned with analyzing Douglas and Miller's 
model of the effects of CAB regulation, and the predictions of this 
model as to what would happen to flight frequency with a lowering of 
fares. The second section is similarly concerned with Douglas and Mil- 
ler's work, but the concern there is with their empirical estimates of 
airline costs, the bias which may exist in these estimates, and the im- 
plications of these biases for any calculations of welfare loss from 
CAB regulation. 

Tfie third section presents calculations for some new estimates of 
the impact of CAB regulation on air coach fares for a "typical" domes- 
tic trunk route. The fourth section presents a similar analysis of the 
effects of CAB regulation on coach service quality. The fifth section 
presents an estimate of the amount of cross-subsidy from coach to first 
class, as of 1974. 

fare regulation, flight frequency, and service quality 

Since fares are set exogenously for regulated airlines, it follows 
that if they are to compete, they must do so in other dimensions. Per- 
haps the most important Avay in which they compete is through fliglit 
frequency and seat availability as stated m the text. This service 
rivalry is analyzed rigorously by Douglas and Miller. Through the use 
of comparative static analysis, they derive the relationship between 
the overall fare level, as set by the CAB, and the amount of capacity 
which will be provided.^^^ 

Their analysis of this issue can be summarized as follows : to sim- 
plify, they assume that airline competition goes on on a single route, 
for which the quantity of people wishing to fly is Q, the (regulated) 
fare level is P, and the total capacity provided is S (which can be 
interpreted to be either flights or seats ; they assume plane size constant 
for this model ) . With non-collusive service quality competition, excess 
profits in the industry will be competed to zero. Hence, 

7r = PQ-cQ-hS=0 


205 Economic Regulation of Domestic Air Transport, chapter 4, and Appendix to 
chapter 4. 


where c is the cost per passenger varying only with passengers carried, 
and k is the capacity-related cost per seat. They then differentiate (1) 
implicitly, finding the relationship between regulated fare and capacity 
offered ; it is 


where e^ is the elasticity of demand with respect to fare. If 
dQ/dS<Q/S (a reasonable assumption, equivalent to a downward- 
sloping demand curve in price), and given tlie relationship in (1), 
Douglas and Miller show that dS/dP<0 if and onlv if 

Douglas and ^filler then present empirical evidence tliat c/P ranges 
somewhere between .201 and .'23G, so that dS/sP = for some value of 
ed between —1.36 and —1.25. They then argue tliat cross-section de- 
mand studies of DeVany (1974) and Brown and Watkins (1971) show 
that the demand elasticity for trunk air transportation ranges between 
— 1.1 and —1.3. Thus, Douglas and Miller conclude that demand elasti- 
city is currently in the range which maximizes the amount of capacity 

This argument is an appealing one, but it has several difficulties. 
First, it is Jiot inmiediately clear from the analysis how "flat" the rela- 
tionship between fare and capacity offered is. If the demand curA'e had 
a constant elasticity over a wide range, there would be no maximum 
(indeed, if the demand curve did not contain a point with an elasticity 
of —1.1 to —1.3. Douglas and Miller's formula would not indicate the 
maxinuun directly). 

Second, because our knowledge of the demand elasticity is shakey, 
and because the elasticity is likely to vary fiom route to route, it is 
very difficult to generalize about what fare level maximizes capacity 

Third, while their analysis is ingenious conce])tually, it does not 
take account of such factors as the first-class-coach price discrimina- 
tion which currently CAB regulation allows, or of the peakload pric- 
ing which o])tii7iized regulation might allow. 

It is impossible, given existing data, to take account of all these 
considerations, and as a result is quite ambiguous whether a lowering 
of fares (especially coach fares) would raise or lower tlie total derived 
demand for aircrjift ca])acity. Our a])]:)roach here will be to do a rough 
numerical calculation of the impact of deregulation on both fares and 
service quality, usinir data for a "typical'' domestic trunk route (de- 
scribed beloAv). While this estimate still will have some of the weak- 
nesses mentioned above, it should give a clearer idea as to the likely 
impact of a fare cut on capacity provided. 

Before doing these calculations, however, it is necessary to consider 
some of Douglas and Miller's empirical calculations which I'elate to 
optimal fares and service quality : their estimates of trunk airline costs. 

COST esti:mates 

There is evidence indicating that Douglas and Miller's cost esti- 
mates are upward-biased. Consider the case of the Los Angeles-San 


Francisco route in California, a run of 347 miles most commonly per- 
formed by a Boeing 727-200. For the sake of argument, let us assume 
that the carriers on this route managed to achieve a 100 per cent load 
factor. Douglas and Miller's cost equation for the 727-200 would then 
predict a total cost (including indirect expenses and return on invest- 
ment) of $17.47 per passenger.^"® This compares with an actual 1971 
fare of $14.81.^°^ If Douglas and Miller's cost equation were correct, 
therefore, it would imply that California fares were set to yield a loss 
of $2.66 per passenger at 100 per cent load factor. The 100 per cent 
load factor was, of course, chosen for illustrative purposes only — it 
is neither feasible nor desirable in scheduled service. With a more 
reasonable load factor, say, seventy per cent, the cost per passenger 
carried in the Los Angeles-San Francisco route would be $21.20 per 
passenger carried,^^® implying a loss of $6.39 per passenger, or 43.2 
per cent of the fare actually collected. Given that Pacific Southwest 
Airlines, operating on that and other routes with similar fare yields, 
earned a positive profit on that route in 1971,^°® there is something 
implausible about Douglas and Miller's costs. A similar degree of 
implausibility would be found if their equation were used to pi*edict 
fares in any of the other California intrastate markets.^^° 

There are several ways of explaining this discrepancy. The first is 
that there is some great, inherent difference between the costs of the 
California intrastate carriers and the domestic trunks. There is evi- 
dence against this hypothesis. First, as Douglas and Miller point out, 
all domestic trunk carriers certificated to fly on the intrastate routes 
in California have chosen to compete with the intrastate carriers 
rather intensively at the low California fares, rather than abandon 
tlie markets to ''lower cost" comj^etition. This has been a long-run 
decision, since PSA has been charging the low fares and holding a 
large market share ever since the mid-1960's. Under these circum- 
stances, the managements of the trunk carriers would have to be non- 
profit maximizers of the worst sort to invest as heavily as they have 
to compete in these markets at such low fares if indeed they would in 
doing so lose the sort of money suggested by Douglas and Miller's co^^t 
equations. The second reason to make one doubt that the interstate 
carriers have costs si^ificantly higher than those of the intrastate 
carriers is that a previously-estimated set of cost equations, based on 
domestic trunk carrier data, predicted (as of 1968) the Los Angeles- 
San Francisco and Los Angeles-San Diego fares to within three per- 
cent, assuming a more reasonable load factor of sixty per cent and a 
normal after-tax profit of 7.5 per cent.^^^ 

If the notion can be safely rejected that the trunk carriers have costs 
far too high to break even at the fares in California, then what ex- 
plains the high Douglas-Miller cost estimates? Another possibility, 
briefly suggested by Douglas and Miller, is that the quality of service 

»* These calculations are based on ibid., Table 2.7, p. 24. 

^•^■^ Keeler, Resource Allocation in Intercity Tassenger Transportation, p. 112. 

^o** Tills is again based on the cost estimates presented in Douglas and Miller, Table 2.7, 
p. 24. 

-'^'^ In fact, even after substantial losses from non-airline operations, PSA earned a 7.1 
percent return on equity In 1971. See Moody's Transportation Manual, 1973, pp. 1358- 

aio por a complete discussion of the California fare yields, see Slmat, et al. vol. II, 
chapter II. 

^Keeler (1972), p. 417. 


provided in coach service on interstate routes (and hence implicit in 
their costs) is significantly higher than on California intrastate routes. 
In making the above comparison, we have already adjusted tor the 
most important quality variable, load factor, and found that that 
cannot explain the observed discrepancies. What of other possible 
quality differences? Reservation service is similar on the two airlines. 
Baggage service is similar, as well. Two aspects of service quality^ 
however, deserve closer attention, food service and seating density. 
On the first count, it is true that the intrastate carriers in California 
and Texas do not serve meals. But then, neither do the domestic trunk 
carriers on all but an infinitestimal fraction of their flights traveling 
for distances under 350 miles. A properly-estimated cost function for 
the domestic trunk airlines would allow passenger service costs to rise 
with distance, so that meals would be implicitly included only as 
flights got longer (Keeler, 1972, attempts to do this). Finally, regard- 
ing seating density, it can be shown that PSA's Boeing 727-200 air- 
craft, seating 158 passengers, have a seat pitch of thirty-four inches, 
and a seat width of eighteen inches, identical to the figures which 
Douglas and Miller assume in their calculations. ^^ In short, it is 
difficult to sec what quality difference Douglas and Miller might be 
talking about between the Los Angeles-San Francisco run and equiva- 
lent interstate runs elsewhere.^^^ 

What are the reasons for Douglas and Miller's high cost estimates? 
There are several sources of upward bias, and it is not the aim of this 
piece to run through them all. Some, however, related to the capital 
costs of aircraft operation, are especially worthy of notice. First, con- 
sider aircraft equipment acquisition costs. To a one, the costs listed 
and assumed by Douglas and Miller for 1971 are higher than airlines 
are on record as paying in the year 197 Jf,, Consider the case of Boeing 
737-200. Douglas and Miller list a purchase price of $6.0 million as of 
1971. Yet, in 1974, Southwest Airlines paid only $5.25 million for a 
new 737-200. This lower price is confirmed by the Airliner Price List 
in Flight International as of October, 1974.^^* Or, consider the 727-200. 
Confidential responses to a Senate questionnaire reveal that a major 
airline paid $7.7 million per new 727-200 ordered in 1974. The price 
list in Flight International puts the price as of late 1974 at $8.0-8.4 
million. Douglas and Miller assume a 1971 price of $8.4 million. Or, 
yet again, consider the case of the Douglas DC-10. Flight Interna- 
tional puts the 1974 price at $21 million or more apiece as of late 1974. 
Yet Douglas and Miller list a 1971 price of $23 million. 

Consider now the issue of seating capacities. For the 727-200, Doug- 
las and Miller assume a seating capacity of 148 in an all-coach con- 
figuration, with eighteen inch seat width and thirty-four inch seat 
pitch (no course is given to back this up). Yet builder's diagrams 
would seem to indicate that with this seat size, the 727-200 will seat 
163, including galley space.^^^ 

2^ In fact, a 727-200 with a small galley will hold up to 163 passengers with seats of 
this size in an all-coach configuration. See Boeing (1972), p. 18. Nevertheless, we allow 
153 seats in our subsequent welfare loss calculations to assume adequate galley space. 

^=' Douglas and Miller's cost model would have equal difficulty predicting interstate fares 
In Texas. See SImat, et al. (1976), chapter III. 

214 Southwest Airlines Annual Report, 1973. and Flight International 106 (Oct. 31» 
1974). pp. 590-591. 

215 Boeing Aircraft Co., Manual D6-58324 (Seattle, 1972), p. 18. 


For the Douglas DC-10 and the same seat size, the capacity as listed 
in builder's diagrams is 330, and yet Douglas and Miller assume a 
capacity of 280, again with no explanation.^^^ 

Consider next the matter of aircraft utilization rates. For every 
aircraft type, Douglas and Miller assume an annual utilization rate of 
3.000 block hours.-^^ Yet, as of the late 1960's, every aircraft type used 
in their final cost calculations was utilized more intensively than this 
by the domestic trunk carriers.^^^ (It may well be that utilization rates 
were lower for some of these aircraft in the 1970's but this is likely 
to be brought about by the overinvestment induced by high, cartel- 
level fares of the CAB, and should therefore not be built into estimates 
of optimal fares.) 

Finally, consider the matter of return on investment. Douglas and 
Miller would seem to have built into their calculations an after-tax 
return on investment of 12 percent. In their cost calculation, they 
state the disclaimer that this return is that specified by the CAB, and 
that they do not necessarily endorse it. But if that is the case, should 
estimated optimal fares be based on such a return, when in fact in- 
vestment has flowed into the airline industry over the past forty years, 
so as to drive the return on investment consistently down to 7.5 
percent ? ^^^ 

No claim is made here that a firm with more efficiency (or a better 
sales force or whatever) than average cannot make more than 7.5 per- 
cent return. There is plenty of evidence that some firms, both regu- 
lated and unregulated, can do so. But it is difficult to see what evi- 
dence there is to support use of a twelve percent pre-tax return on in- 
vestment for a firm with typical costs and selling success, given that 
such firms have survived for nearly forty years now on an average 
return of 7.5 percent after taxes. 

The upshot of this brief discussion of costs is that it might be worth- 
while to recalculate some of the Douglas-Miller results on the basis 
of another set of cost estimates, which, it can be argued, represent 
more accurately what could be achieved with the looser sort of airline 
regulation which has been in effect within the states of California and 
Texas. Previous work of the present author (Keeler, 1972) provides 
such cost estimates. The model used in that paper, while calibrated 
with trunk data, came within three percent of predicting California 
intrastate fares after adjustment for the higher load factors achievable 
with deregulation. 

How can we expect deregulation to affect load factor ? Furthermore, 
can we expect the load factor generated by unregulated markets to be 
an optimal one ? 

In Keeler (1972), it was assumed that deregulated carriers would 
achieve a load factor of sixty per cent. Fortuitously, this is almost 
exactly the load factor which prevailed on the Los Angeles-San Fran- 

^1" Fliijht International 98 (Nov. 18, 1970). 

21' Douglas and Miller, p. 23. 

218 Keeler (1972), p. 405. Also, it is worth noting that even during the 1974-75 slump, 
Braniff, a very extensive user of 727-200's, managed to achieve a block utilization rate 
of over nine hours per day or 3285 per year. See Aviation Week 100 (Oct. 28, 1975), p. 45. 

219 Keeler (1971), p. 19-23. 


Cisco and Los Angeles-Oakland run in the late 1960's and early 
1970's.-2" and it also turns out to be the load factor Douo-las and Miller 
find optimal for a ''typical" route.--^ :\roreover, a recent doctoral dis- 
sertation by Dorman has demonstrated theoretically that under a very 
plausible s'et of assumptions, all-out competition in the price and 
service-quality dimensions in air transportation should indeed gen- 
erate an optimal amount of each, at least on higher-density routes.-^ 
On the basis of this evidence, we shall assume that with deregulation, 
aircraft will be operated in an all-coach configuration at a sixty per 
cent load factor. This is a rough assumption, but it is suggestive of 
the price and service-quality combination which deregulated markets 
seem likely to achieve on the average. 


For the following calculations, we use data for a "typical" domestic 
trunk airline trip, in that its characteristics in terms of length, num- 
ber of stops, and passenger traffic density are the same as those for 
interstate typical trunk airline passenger as of the year 1974. Such a 
trip, on a CxVB-regulated trunk carrier, had a length of haul of 794.7 
miles, is ser^^ed by an aircraft with seating for 137 passengers, of 
which 22 scats wore first class, and ll.") were coach.--' The average staire 
length was 582.2 miles, implying that the aircraft made an intermedi- 
ate stop 36.5 per cent of the time (794.7/582.2 = 1.365). This sort of 
service is that most likely to be provided by an aircraft of intermedi- 
ate range, such as the Boeing 727-200. Because this aircraft is a coni- 
monly used one on both inter- and intrastate routes, and because it is 
well-suited to such a "typical" trip, we assume, for purposes of cost 
calculation, that this is the aircraft which would be used for such a 
route were it deregulated. The 1974 costs of operating such an aircraft 
are shown in Table Al, based on data from the CAB and other 

Indirect operating cost data were updated from 1968 to 1974 in the 
following manner: the equation from the 1972 paper was used first to 
predict total indirect costs per airline, and it was found that 1974 
costs were 1.474 times the 1968 ])i'ediction.-'-"* Therefore, all cost co- 

--'^ PSA's load factor on the Los Antrolos-San Franrisro ronto in lOflS was fifty fivo ppr- 
cent : In 1972. the other year for which calculations are made here. It was sixty percent. 
See California Public Utilities Commission, Transportation Division. "Passengers on Board 
and Load Factors for Scheduled Air Carriers on Nonstop Flights between California 
Airports," 1969 and 1972. 

-^ For route density of 400 passengers per day. Douglas and Miller find an optimal 
load factor (on a Boeing 727-200) of fifty-seven percent with a $10 per hour time value, 
and sixty-one percent with a $5 per hour value. For a 1,000 mile haul, they find an 
optimal load factor of sixty to sixty-three percent, again depending on the value of 
schedule delay time. Thus, for the 792 mile haul discussed here sixty percent might seem 
roughly correct. It must be remembered, however, that because the operating cost data 
used here differ from those of Douglas and Miller, optimal load factor calculations based 
on them are likely to be different from those of Douglas and Miller, at least by a small 
amount. Thus, the sixty percent assumed here Is an approximation. Dorman (1976) pro- 
vides a more detailed Integration of the present cost model with the Douglas and Miller 
stocliastic delay model. 

2- On low-density routes, air transportation is subject to the same sort of service-related 
economies of traffic density as pointed out for buses by Mohrlng (1972), and private 
markets will fail as a result ; subsidies are necessary to achieve optima in such markets. 

223 These figures come from the U.S. Civil Aeronautics Board Air Carrier Traffic Statistics 
(December 1975). p. 6. 

22* T. Keeler, prepared statement, Kennedy Hearings, vol. II, pp. 1303. 


efficients were increase by 47.7 per cent. This yields the following in- 
direct cost equation : 

$.259 $.003330 

TCP= (1 + ^)+ Z>+$.0203Z> (3> 


where TCP is total indirect trip cost per passenger, D is trip distance, 
and S is the average number of intermediate stops per trip. At a 
60 per cent load factor, this implies an indirect cost per passenger of 
$21.14 with the values of D and S described above. 

Table A1. Direct operating costs, 727-200, 1972 

Cost per 
Expense Class hlock-fiour 

Cash expenses (flying operations and maintenance)^ $708 

Capital costs ^ 412 

Total cost per block-hour 1, 120 

1 From U.S. Civil Aeronautics Board "Aircraft Operating Cost and Performance Sta- 
tistics" (1972-73) ; figure pertains to all operations of Domestic Trunk Carriers, Boeing 

2 This figure comes from T. Keeler, prepared statement, Kennedy Hearings, Table 3. 
This table, however, was misprinted by the U.S. Government Printing Office. The indirect 
capital cost (ground support capital) for a 727-200 should have been $44 per block-hour. 
(The line containing direct operating costs was incorrectly omitted from Table 3 of my 
prepared statement.) 

From Table Al, we note that the total direct operating cost for a 
Boeing 727-200 is $1,120 per hour. To convert this to the cost for a 
trip of the sort described, we use Douglas and Miller's equation relat- 
ing length of hop and travel time for a 727-200 : 



On the basis of this, a trip of the dimensions described above should 
take 125.5 minutes, or 2,092 hours in block time. Total cost is then 
$2,343 per plane trip with a 153-seat, all-coach configuration (includ- 
ing full galley space), and a 60 per cent load factor; this generates a 
direct cost of $25.52 per passenger trip. Total cost (and hence our 
estimated efficient fare) is then $46.70 per trip, or 5.88 cents per 
passenger-mile. This compares with an actual 1974 coach revenue 
yield of 6.94 cents per passenger-mile. ^^^ Thus, if regulated fares are 
calculated on the basis of revenue yield, the extra fare incurred from 
regulation is about one cent per passenger-mile, implying a markup 
of 18.2 per cent. This means that passengers who are currently travel- 
ing by coach on trunk routes paid extra fares totaling about $1.1 

But use of revenue yields per mile for regulated routes in the pres- 
ent case will clearly yield a downward-biased estimate of the fare costs 
of re<2:ulation. The reason is simple : the cost model from Keeler (1972) 
predicts the unrestricted^ peak-period fare on unregulated routes. The 

225 Ernnnmic Reaulation of Domestic Air Transport, p. 21. 
226rrSC>B. Air- Carrier Traffic Sttatistics (December 1975), p. 6. 

227 ($.0694-0588) X (103.8 billion passenger-miles )!=«$1.1 billion. Coach passenger-mile 
figures are from ibid., p. 6. 


intrastate routes also have discount fares, and other recent work of 
this writer has shown that controlling for distance, the discount and 
off-peak fares on the top three intrastate routes in the country are in 
fact only sixty per cent of the average discount fare on interstate 
loutes (see text section V-B). Thus, if revenue yield is used as a fare 
estimate on the interstate routes, the same should be done for intra- 
state routes. Inasmuch as the cost model used here predicts regular, 
daytime fares on the unregulated routes, it is obvious that the same 
sort of fare should be used for comparison on the interstate ones. Fur- 
thermore, after all the concern shown here about normalizing for serv- 
ice quality, it is clearly inconsistent to compare interstate revenue 
yields with intrastate fares, because much of the interstate revenue 
stems from restricted discount fares, which offer a service quality dis- 
tinctly inferior to that of the normal, peak-period fare. It may well be 
that more such discount fares are available on interstate routes, but 
just the same, it is important to compare the regular intrastate fare 
with the equivalent regular interstate fare to get an alternative (and 
perhaps upper-bound) estimate of the impact of regulation on fares. 
On a typical route of 795 miles, the CAB fare structure in 1974 would 
generate a regular coach fare yield of about 8.5 cents per mile.^^^ On 
this basis, excess fares paid coach passengers from regulation were 
$2.7 billion. [$.085-.0588) X 103.8 billion passengers-$2.7= billion]. 


Let us now recapitulate our assumptions about the changes in fares 
and service quality in coach service before and after deregulation: be- 
fore deregulation, we assume a route of 400 passengers as described in 
the text. Overall, 1974 coach passengor-miles were 88.3 per cent of the 
total,229 so we assume 353 of the 400 traveled in coach. Now, in 1974, 
the overall load factor in coach and first class was 55.7 per cent, corre- 
sponding to 58.9 per cent in coach and 39.7 per cent in first-class. With 
a plane capacity of 137, this implies that each plane carries 76.3 pas- 
sengers per flight, and hence with a route density of 400 passengers 
per day, a total of 5.24 flights per day in both directions combined. 

To analyze the effects of regulatory policy on interstate coach pas- 
sengers, we assume that for them in 1974 the average plane size was 
115 passengers, and that if the coach compartment is full on a given 
flight, they will wait till the next one rather than book first-class. 
Thus, the effective plane size is 115 seats, the number of passengers in 
coach is G7.7 per flight, and the flight frequency is 5.25 planes daily 
in both directions combined. 

We assume that with deregulation, each plane will seat 153 passen- 
gers, and that fares will be set to yield a normal profit at a 60 per cent 
load factor; so after deregulation, each flight will contain 91.8 passen- 
gers (153 X .6). But that does not tell us enough to determine the 
flight frequency. The lower fares from deregulation will induce extra 
traffic. It has already has argued that the Brown- Watkins DeVany 
demand elasticity estimates for coach service have taken into account 
the change in service quality which will occur when a fare changes, 

228 por oxamplp. the rejrnlar Atlanta-New York fare In 1974 was $64.80 for 755 miles, 
or 8.6 cents per mile. On the other hand, the Boston-Chlcap:o fare was $71.30, or 8.3 cents 
per mile. 1974 fares are from the fficial Airline Guide (Oct. 1, 1974). 

"•« USCAB, Air Carrier Traffic Statistics (December 1975), p. 6. 


in order to keep profits at a normal leA'Cl. Thus, to determine the im- 
pact of deregulation on route density, we use the average of the De- 
Vany and Brown-Watkins demand elasticities. The average value of 
their estimates is —1.19.^^° To get a conservative estimate of the service 
costs of CAB regulation, we shall assume that the relevant price 
change occurring from deregulation is from the average revenue yield 
under regulation to the regular fare under deregulation, so that the 
one- way fare per mile will decline from 6.94 cents before deregulation 
to 5.88 cents after. Applying the demand elasticity of — 1.19 to this, we 
find that total coach passengers on the route increase from 353 to 484. 
For the sake of conservatism, we shall assume that first-class passen- 
gers moving to coach are included in this figure, or that they do not 
fly as a result of lack of first-class service. Thus, after deregulation, 
total flights are 484/91.8 = 5.27, and the load factor, to repeat, is sixty 
per cent. 

No claim is m.ade here of perfect accuracy in any of these figures. 
They are obviously assumptions, and the only claim for them is that 
they are in many ways more plausible and logically consistent than 
those of Douglas and Miller. We now have estimates of all the needed 
parameters to feed pre- and post-deregulation data into Douglas and 
Miller's estimated delay function, and come up with estimates of total 
schedule delay per passenger both before and after deregulation. 

The frequency delay function is stated quite clearly by Douglas and 
Miller; it is Z> = 92/^"-*^^ where DF is frequency delay time in minutes 
per passenger, and F is flight frequency, measured in flights in both 
directions per day.^^^ The stochastic delay function is nowhere explic- 
itly stated, but with some effort, it is possible to infer it from what is 
written in the book.^^^ It is 

Z>, = 14.3 N^^-^'^ {B^-N^Y^-'^ (1790/7^) (4) 

where Dg is stochastic delay per passenger in minutes, Sf is average 
plane seating capacity, N is the average number of seats per flight, and 
F is flight frequency, as before. 

Table A2 states the assumptions made regarding each of the param- 
eters of the delay function, and also summarizes the results regard- 
ing stochastic delay. The results are surprising: they indicate that 
overall schedule delay actually goes doion after deregulation, from 98 
to 84 minutes per passenger-trip. This would seem like a paradoxical 
conclusion, because the load factor has gone up as a result of deregula- 
tion. However, a careful look at the stochastic delay function indicates 
that it is the absolute size of the difference between number of pas- 
sengers per seating capacity that determines stochastic delay, and not 
the ratio. And this size difference clearly increases after deregulation. 
Furthermore, demand elasticity appears to be great enough to keep 
flight frequency more or less unchanged. At a $5 per hour value of 
schedule delay time, the value of service improvements to coach pas- 
senger would be about $130 million.^^^ 

230 DoVany's estimate is —1.07 ; Brown and Watkins' estimate is —1.3. See Douglas 
and Miller (1974), p. 37. 

^ lUd., p. 83. 

232 ThifL, pp. 82-108. panftim,. 

2-'''' There were approximately 131 million coach trips on domestic trunk carriers in this 
country in 1974 and savings from above are 12 minutes, or $1.00 per trip. USCAB, De- 
cember, 1975, op. cit., p. 6 































With Without 

Assumption regulation regulation 

A. Assuming regulated fare is revenue yield: 

Average coach trips/day 

Average fare per mile (cents) 

Average coach load factor (percent) 

Average lst-jlass lead factor (percent) 

Average coach capacity per plane 

Average number of flights per day 

Coach passengers per plane 

Frequency delay per passenger (minutes) 43.2 43. 1 

Stochastic delay per passenger (minutes) 54.8 40.9 

Total schedule delay per passenger (minutes) 98.0 84.0 

B. Assuming regulated fare is official undiscounted fare: 

Average coach trips per day 

Average fare per mile (cents)- 

Coach load factor (percent) 

Coach capacity per plane 

Flights per day 

Coach passengers per plane 

Schedule delay per passenger (minutes) 

Obviously, the roughness of the assumptions behind these calcula- 
tions makes the observed difference far too low to be of any signifi- 
cance. We can conclude, however, that on the average, regulation has 
had little effect on average schedule delay. (It may have some effect 
on such things as seat size and meal service ; but in the case of meal- 
service, our cost model is based on domestic trunk data, and makes a 
more than adequate allowance for passenger meals, one which in- 
creases linearly with length of haul.) 

All calculations so far have been based on the assumption that the 
appropriate coach fare on interstate routes should be represented 
by average revenue yield per passenger. Suppose we now base our 
calculations on the assumption that actual fare yields are more accu- 
rate. Under these circumstances, the observed price cut from deregu- 
lation will be substantially greater than implied by previous figures, 
as shown under heading B in Table A2. It will be 30.8 per cent, from 
8.5 cents to 5.88 cents per mile on the typical route. This implies a more 
dramatic increase in passengers as a*^ result of deregulation (to 550 
per day) and an increase in flight frequency to 5.99 flights per day. 
Average schedule delay per passeger would drop from 98 minutes 
with regulation to 76.7 without. The details of the assumptions and 
calculations are again shown in Table A2. At a $5 per hour time value, 
this would imply a total of $230 million worth of service quality 
improvement for coach passengers as a result of deregulation.^^* 


It is instructive to make an estimate of the first-class loss per 
passenger-mile carried, using the cost model developed earlier. It is 
possible to revise the cost model in the Appendix to get a rough esti- 
mate of the first-class costs by noting the following facts: a first-class 
seat takes up 1.68 times as much space as a coach seat (they are four 

"■" Baspd on the same lisrures as footnote 233, but with a time saving of 21 minutes per 
passenger with deregulation. 


abreast, whereas coach is six abreast in a typical airliner, and first- 
class seats tend to be 38 inches in pitchy compared with 36 in coach). 
Furthermore, the first-class load factor m 1974 was only 39.7 per cent. 
Taking account of these facts, the indirect cost equation (2) of this 
Appendix, should be revised to read as follows : 

' ^^^=^(l+^)(l-e8)+ $-QQ3^|g^(l-^8) +$.0203 

where S is the average number of stops in an average trip (found to 
be 1.35) and D is the average trip length, found to be 795 miles in the 
Appendix for 1974. Thus, the first-class indirect cost per trip is $28.83 
per trip. Since practically all direct costs are dependent on airliner 
"Capacity offered, the direct cost of first-class is simply the coach cost 
per seat-trip times 1.68, divided by .397. The direct first-class cost per 
trip is thus ($15.31) X (1.68)/(.397) =$64.79. (The cost per coach seat- 
trip is $25.52 X .6, from p. 15 of the Appendix, given our assumption 
•of a 60 per cent load factor). The total cost of a 795-mile first-class 
trip is therefore $28.83 +$64.79 =$93.62, or 11.8 cents per mile. On the 
other hand, the first-class revenue yield was only 9.49 cents per mile 
in 1974.235 

235 Calculated from USCAB, Air Carrier Financial Statistics (Fourth Quarter, 1975), p. 3, 
and USCAB Air Carrier Traffic Statistics (December, 1975), p. 6. 


(By Ricjiard Zeckiiauser and Albert Nichols,* 
John F. Kennedy School of Government, Harvard University) 

*The authors wish to thank Mary Jane Bolle, Stephen Breyer, Jonathan Brock, Nancy 
Jackson, Michael Klass. John Mendeloff. Ro^er Noll. Leonard Weiss, and two anonymous 
reviewers for helpful comments. Jack Goldstone provided most able researcli assistance. 



Coiifrress j^assed tlie Occupational Safety and Health Act 
(DSHAct) of 1970 "to assure so far as possible every workin;nr man 
and woman in the Xation safe and healthful working: conditions.'' The 
performance of the Occupational Safety and Health Administration 
(OSHA). created to carry out most of the act's provisions, has been 
at best a disappointment. Industry observers view it as another costly, 
unjustified goveriunent intrusion, wliile labor decries its failure to 
achieve noticeable reductions in the toll taken by work-related injuries 
and illnesses. 

This overview study identifies the competing considerations in the 
formulation of government policy for occupational safety and health. 
It then evaluates the effectiveness of current and prospective regula- 
tory efforts and (with appropriate qualifications, given the investiga- 
tion's limited nature) nuikes recounnendations on profitable directions 
for reform. 

occurational safety and health : 3tarket performance and the 

goverxmext's role 

Safety and Health Determination m Perfectly Competitive Markets 
The traditional economic model of safety and health determination 
in perfectly competitive mai'kets assumes that workers and employers 
are well informed alx)ut levels of risk in the workplace. In that model, 
workers demand higher wages if their emi)loyment exposes them to 
risk. A firm determines the levels of safety and health to be offered 
by balancing the cost of achieving higher occupational safety and 
health (OSII) levels against its savinofs in the costs of accidents and 
illnesses and in wages paid for assuming risk. Together, the self-in- 
terested actions of workers and employers lead to an efficient outcome, 
one that minimizes the sum of losses due to accidents and illnesses and 
the costs of prevention. 

Apjyropriate Levels of Safety and Health 

"Safe and healthful working conditions'- cannot be defined in any 
absolute sense; short of prohibiting all work, some level of occupa- 
tional risk will always be present. Different levels of risk must be 
judged both as to their efficiency and their equity. The competitive 
market outcome satisfies the efficiency criterion. Levels of health and 
safety, however, will be distributed unequally; in general, poor people, 
who will accept a smaller monetary premium to incur risk, will be 
exposed to greater risks. But efforts to redress these perceived inequi- 
ties may actually diminish the welfare of the poor by denying them 
certain employment opportunities. 



Marhet Imperfections 

Eeal labor markets diverge significantly from the theoretical model 
of perfect competition. Workers and firms do not possess perfect in- 
formation, the incentives for market provision of such information 
being weak. The limited role of experience rating in workmen's com- 
pensatian systems, and the presence of significant governmental in- 
come support programs and subsidies for health care, generate finan- 
cial externalities; workers and firms do not bear the full costs of 
decisions to incur OSH risks. Such externalities offer perhaps the 
strongest argument for consideration of government interventions to 
increase levels of occupational safety and health. 

Regulatory Intervention 

Government interventions should be designed to provide maximum 
OSH gains for the resource expenditures they entail, as well as to 
induce a level of OSH-related expenditures that reflects the prefer- 
ences of workers, employers, and society at large. The rationale for 
governmental provision of information is clear and relatively uncon- 
troversial. So too, it seems evident that the government should seek 
to encourage, not impede, the functioning of existing mechanisms to 
promote OSH, predominantly the market and workmen's 

Direct regulation of working conditions through standards is an 
approach worthy of consideration as one response to financial exter- 
nalities and imperfect information. But only a performance assess- 
ment can determine whether a standard system is justified in practice. 
Physical standards bear on only a small proportion of the factors 
contributing to OSH levels. Furthermore, their inflexibility is an im- 
portant drawback when they are applied across firms and industries 
with significantly different costs and benefits. Given the limitations of 
a standards system, alternative approaches, including most partic- 
ularly those that rely on incentives, should receive careful 


Government has long been involved with OSH, through both direct 
regulation and workmen's compensation. The OSHAct of 1970 grew 
out of the concern that OSH losses "impose a substantial burden upon 
. . . interstate commerce in terms of lost production, wage loss, medical 
expenses, and disability compensation payments." The objective was 
to reduce injuries and illnesses in the workplace. The major tool pro- 
vided by the act was the authority to promulgate and enforce stand- 
ards. States that develop plans which are "at least as effective" as 
OSHA's may regain responsibility for OSH regulation. 



OSHA is an assistant secretaryship in the Department of Labor. 
It has more than 2,700 employees and a fiscal year 1977 budget of 
about $130 million. In addition to its own standard-setting and en- 
forcement activities, OSHA is responsible for monitoring 24 State 
plans approved under section 18(b) of the OSHAct. 


OSHA initially adopted more than 4.000 standards from pre- 
existing Federal agencies and national OSPI "consensus organiza- 
tions," but has promulgated only a handful of new permanent stand- 
ards in the 7 years of its existence. The standards-settmg process is 
now accelerating somewhat, particularly in health. Standards have 
been criticized on many grounds, including unintelligibility and then- 
failure to address significant OSH problems. OSHA has not consid- 
ered economic costs explicity in the standards-settmg process. 
"Feasibility" has been interpreted primarily as technical feasibility, 
with little attention to the balance of costs and benefits. 

OSHA's first inspection priorities are employee complaints of im- 
minent danger and investigations of catastrophes. General schedule 
inspections, in an effort to increase their productivity, are now being 
targeted at industries on a "worst-first" basis. OSHA inspectors visit 
only about 2 percent of the covered workplaces each year. Voluntary 
compliance is low ; only about 20 percent of the firms inspected are 
"in compliance." Fines for violations are very low, averaging less 
than $16o per cited firm. Penalties for failure to abate a cited viola- 
tion and for repeated violations, however, are significant. 


Although the data on OSHA's effectiveness are too sketchy to per- 
mit precise estimates, it is clear that OSHA's impact on injuries has 
been minimal. Most studies have been unable to find any statistically 
significant improvement. The most optimistic estimate, based on a 
study of "preventable" accidents in California, suggests a 2 to 3 per- 
cent reduction in injuries and a 5 percent decline in deaths. There is 
no evidence that injury rates decrease after a firm is inspected. No 
reliable data are available on OSHA-induced trends in occupational 
illnesses, but there is little reason to believe that there have been signifi- 
cant improvements given the limited attention devoted to health 
hazards by OSHA in the past. 

The major costs imposed by OSHA are those associated with com- 
pliance: purchases of required new equipment and modifications of 
e?:isting equipment and plants, and losses of productivity through re- 
quired changes in work practices. Analyses of specific standards yield 
annual cost estimates of about $110 million for full compliance with 
the proposed inorganic arsenic standard and between $241 million 
and $1.3 billion for the coke oven emissions standard. By far the most 
costly standard is noise ; moving to an 85 dBA standard would mean 
about $18.5 billion in capital costs alone over 5 years, for an incre- 
mental cost of $8 billion over tlie existing 90 dBA standard. In con- 
trast, fines are a small burden, totaling less than $10 million in 1975. 


Well documented research results are not available on most aspects 
of OSHA's performance, indeed on many critical issues in the OSH 
area. Only limited time and resources were available for this study. 
Our recommendations can only point to potentially useful areas for 
change; they are not intended as a blueprint for reform. 

The Role of OSHA 

OSHA should be assessed as a complement to existing market and 
gov^ernmental mechanisms for promoting OSH. Decentralized mech- 

83-944—78 12 


anisms have strong theoretical merit. Efforts to increase the incen- 
tives for firms to provide safer and more healthful working condi- 
tions — including changes in workmen's compensation laws to tie 
firms' costs more closely to their individual records, and the possible 
imposition of taxes on injuries or hazardous conditions — should make 
up a major portion of any governmental effort to improve OSH. 

OSHA has overstressed worker safety as opposed to health, al- 
though it is now beginning to redress that imbalance. The justifica- 
tions for Government intervention, particularly for direct regulation 
through standards, are much stronger for health, where information 
is more difficult for workers to obtain and to assess, and where exter- 
nalities are greater. However, health standards have the potential to 
impose far higher costs as well; they should be formulated with a 
careful and explicit consideration of economic factors. 

Reform of OSHA 

Under tlie direction of Dr. Morton Corn, OSHA undertook a number 
of beneficial, incremental reforms. Some of the nuisance standards 
adopted from consensus organizations were eliminated or rewritten. 
Performance standards received more consideration. General schedule 
inspections were downgraded in priority to achieve greater produc- 
tivity. Experimental new regulatory approaches were formulated in 
cooperation with management and labor in a demonstration program 
directed to the high-risk foundry industry. 

Since Dr. Eula Bingham has headed OSHA from the outset of the 
Carter Administration, the agency has embarked on two major initia- 
tives. First, it has proposed the elimination of a large number of 
nuisance standards, many of which had subjected the agency to ridi- 
cule. Second, it has placed a substantially increased emphasis on health- 
related issues, including a proposal for dealing with suspected carcino- 
gens on a generic basis. The generic approach classifies a substance for 
purposes of regulation solely on the basis of the number and type of 
different tests showing its potential carcinogenicity. Highly relevant 
considerations, such as the potency of the substance as a carcinogen, the 
cost of reducing exposure to it, and the aA^ailability of substitute sub- 
stances are considered neither in the classification scheme nor in the 
means or level to which it should be controlled and monitored. OSHA 
has recently proposed a permanent standard for reducing worker ex- 
posure to benzene. That standard too was formulated without taking 
account of either the costs of reducing exposure to benzene or the 
health benefits that such a reduction would offer. Fully two thirds of 
the operating costs for the benzene standard derive from monitoring 
and surveillance requirements, yet nowhere is any assessment made 
of the benefits of such activities, particularly in comparison with 
reduced exposure itself. 

The major continuing deficiencv of OSHA's standard-setting pro- 
cedures, in both the present and former administrations, is the exclu- 
sion of economic considerations, except in the most extreme cases where 
ihf". viability of an industry is in question. Although assessment 
of safety and health benefits in dollar terms would pose extreme diffi- 
culties, both political and analytic, cost-effectiveness analysis should 
be used to highlight the trade-offs implicit in different standards. 
Explicit consideration of economic costs would probably require writ- 
ing OSHA's legislative mandate. 


OSHA should facilitate market function by expanding its efforts to 
provide information to both workers and employers. Onsite con- 
sultation, which would require statuatory change, would be a useful, 
if predominantly symbolic, step in this direction. It would help as 
well by casting OSHA in a cooperative rather than an adversary 

To redirect its efforts in more productive ways, OSHA must im- 
prove its ability to collect and to assess data relating working condi- 
tions, OSPI risks, regulatory interventions, and economic costs. An 
enhanced policy analysis capability in the agency will be required. 

The division of responsibilities between State and Federal govern- 
ments, though a delicate political issue, should be examined. A move 
toward either extreme might be preferable to the present mixed sys- 
tem, particularly given the difficulties of determining whether State- 
run programs meet the requirement that they he "at least as eff'ective" 
as Federal efforts. 

Various proposals have been advanced to restructure OSHA and re- 
lated organizations. They include making OSHA an independent 
regulatory agency and shifting the National Institute for Occupa- 
tional Safety and Health (XIOSH) into the Department of Labor, 
with or under OSHA. Some such reorganizational efforts should per- 
haps be considered, though to undertake them would drain political 
energies that could be directed to more demonstrably beneficial modes 
of reform. 

Concluding Rernarks 

OSHA has been a disappointment to its supporters, though, given 
the task and its methods, at best weak gains for OSII should liave 
been expected. Business antagonism toward OSHA's poorly directed 
efforts has been reinforced by a growing general citizen disillusion- 
ment with government intervention. OSHA has made some limit-ed 
moves in ap])ropriate directions, particularly in relation to safetv 
problems. However, it has made no attempt to* rationalize its interven- 
tions in the potentially nmch more consequential health area, where it 
is rapidly expanding its efforts. OSHA shows no sign that it wishes to 
formulate its policies in a manner that pays careful attention to all 
their consequences. If sufficient l>eneficial reform is to be achieved. 
Congress will have to relegislate OSHA's mandate. The agencv should 
be required to channel its efforts where they can be shown to' be most 
productive, achieving the greatest health gains for whatever resource 
costs thev entail. 


(By Richard Zfxkhauser and Albert Nichol, John F. Kenr^dy 
School of Government, Harvard University) 


Congress passed tlie Occupational Safety and Health Act 
(OSHAct) of 1970 intendincT "to assure so far as possible every work- 
infj man and woman in the Nation safe and healthful working condi- 
tions." ^ Several new agencies were created to carry out the provisions 
of the act, with the major responsibilities falling upon the Occupa- 
tional Safety and Health Administration (OSIIA) within the De- 
partment of Labor. OSHA's primary activities have been the 
promulgation and enforcement of several thousand standards cover- 
ing the physical conditions of workplaces. 

The OSHAct was passed during a time of marked enthusiasm for 
the potential role of the Federal Government in increasing levels of 
safety and health in a variety of areas. AVithin a brief period around 
1970, Congress passed new legislation regulating coal mine healtli 
and safety, air pollution, consumer product safety, and automobile 

Despite the optimism with which the agency was launched, few 
observers are happy with OSHA's performance. Industry views it as 
anotlier unjustified and costly iiitrusion. Workei-s, organized labor, 
and other proponents of strong regulation of occupational safety and 
health have been disappointed by OSHA's failure to secure noticeable, 
much less notable, reductions in the toll taken by work-related in- 
juries and illness. Dissatisfaction with OSHA has been reinforced by 
a growing opposition to government regulation of all kinds. 


Given the present antiregulatory mood of much of the Nation, the 
apparent inability of OSIIA to have a significant impact on occupa- 
tional safety and health problems, and the commitment of the 
Carter administration to consider reorganization of the Federal Gov- 
ernment, a dispassionate analysis of OSHA should help chart future 
directions for the Agency. We do not question the significance of the 
OSH issue : Each year on-the-job accidents kill thousands of workers 
and injure millions; occupational diseases probably strike hundreds 
of thousands, resulting in tens of thousands of deaths. Indeed its 

1 Occupational Safety and HeaUh Act of 1970, section 2(b). 



importance reinforces the need to eA^aluate the effectiveness of current 
and potential government regulation in this area, and to highlight 
the competing considerations in the formulation of policy. 

We begin our analysis with a predominantly theoretical discussion 
of the way levels of safety and health would be determined in per- 
fectly competitive markets, and the various criteria by which such 
levels mi'ght be judged as to their appropriateness. We then consider 
the imperfections present in the markets that determine OSH levels. 
Individuals do not possess perfect information about the risks they 
face, and various institutional arrangements impose rigidities and 
create financial externalities which prevent the market from func- 
tioning efficiently. These imperfections suggest several possible modes 
of government intervention, including providing information and 
operating various incentive mechanisms, such as taxes on threats to or 
losses of health, tort law, and workmen's compensation. An alterna- 
tive, complementary intervention, typified by OSHA, is the direct 
regulation of working conditions through standards. 

Section III presents an overview of Government involvement in 
occupational safety and health, covering activities at the Federal, 
State and local levels prior to the OSHAct. and a summary of the 
act's provisions. In section IV we evaluate OSHA's performance to 
date, discussing both the manner in which the act has been imple- 
mented and its impacts on different groups. Although the evidence 
available is incomplete, it suggests that OSHA has imposed signifi- 
cant costs, apparently without yielding commensurate benefits. 

The final section of the paper recommends a number of directions 
for reform. Aside from proposals on specific OSHA procedures, we 
advocate efforts to enhance the performance of OSH-related incentive 
mechanisms, and to reinforce OSHA's current efforts to shift its 
emphasis from safety issues to health hazards. We stress the need 
to incorporate economic costs more fully into the formulation of 
government regulatory efforts. 

A brief postscript discusses several initiatives that OSHA has un- 
dertaken during the past year under Dr. Eula Bingham, whose ap- 
pointment as Assistant Secretary of Labor for Occupational Safety 
and Health had just been announced when the main body of this re- 
port was completed in February, 1977. Many nitpickino: and unpro- 
ductive safety standards are being eliminated, and OSHA is acceler- 
ating its efforts in the health area. This development would be welcome 
if OSHA would guide its health interventions through careful consid- 
eration of their costs and benefits. Dismayingly. OSHA continues to 
refuse to make estimates of either costs or benefits an integral part of 
its standard setting process. 


Government policy for promoting occupational health and safety 
is a critical issue, and a most complex one. It should be dealt with 
neither lightly nor briefly. This analvsis was commissioned to be con- 
ducted over a short period of time, with limited resources, and with an 
explicit understanding that it was to rely on secondary sources. In 
sum, it can provide only an overview, more capable of identif\nng 
poorly functioning systems than of prescribing specific remedies. 


Additional difficulties confronted this study. The published literature 
and available unpublished literature on OSHA is surprisingly brief. 
No reasonably broad, much less definitive studies of OSHA's per- 
fonnance have been conducted. ^Moreover, gaps and inconsistencies in 
the data base make even the most careful of studies somewhat incon- 
clusive. Thus recommendations of necessity rely heavily on logic, 
theory, and indicative data. There are no controlled experiments to 
offer guidance, no substantial data sets from which unambiguous 
inferences can be drawn. 

Although hesitant to advance conclusions and recommendations, we 
felt it essential to proceed beyond a mere review of OSHA's present 
status. At present, no one is truly happy with the Agency. But the dis- 
satisfactions of various parties are quite different, and indeed are 
often in conflict. Concrete proposals may help sort out the differences, 
as well as reveal more clearly our point of view. 

Some critics, particularly those concerned with the legislative man- 
date of OSHA, assert that it has neglected its major responsibilities. 
Other critics suggest that the Agency has been inefficient in carrying 
out its mission. Finally, a number of economists, who appreciate the 
accomplishments of free markets, and industrial leaders, who see 
OSHA as another costly, perhaps pernicious. Government intrusion 
into their activities, feel that the entire conception of a Government 
Agency regulating health and safety within the workplace is mis- 
guided. Wo believe that much disagreement among those groups would 
be eliminated, and indeed a sensible middle ground established, if a 
cai-eful record of the accomplishments and the costs of OSHA could 
be tallied. 

OccuPATioxAL Safety and Health: 
Market Performance and the Government's Role 

safety and health determination in perfectly 

co:mpi:titive markets 

Policymaking in the Ignited States, a society that is supposedly 
dedicated to free enterprise, should start with a recognition of the 
significant accomplishments of competitive free markets. The gov- 
ernment should only consider stepping in when competitive condi- 
tions cannot be maintained — when, as the saying goes, the market 
fails. Some understanding of the appropriate regulatory role for gov- 
ernment is gained by considering the hypothetical situation where the 
market functions ideally.^ 

If .perfectly competitive markets prevail, the health and safety 
characteristics of employment will be determined by the decisions of 
individuals on both sides of the market, workers and employers, with 
regard to the packages of wages and OSH conditions that they will 
accept and offer. 

The Workers Choice: Wage Premiums as Accompaniments to Risk 

The worker is confronted with a nimiber of different employment 

opportunities. The free flow of information assumed in the competi- 

2 More extended and formal treatments of OSH determination In competitive mirkpts 
mav hp found in tlie literature. See. e.jr.. W. Y. Oi "On tlie Efonouiics of Industrial Safetv." 
Law and Contemporary Problems. 38 (Summer-Autumn, 1974). 669-699. For an enfer- 
taininc. but also instructive. "Safety Fable" on this same subject see R. S. Smith, The 
Occunationnl Safety and Health Act (American Enterprise Institute, Washington. D.C.. 
1976), pp. 20-26. 


tive model assures that he shares all available knowledge about occu- 
pational risks. Furthermore, no one else is concerned with his choice 
of employment: there is no other-regarding behavior, nor will any 
outside party gain or lose resources depending on the worker's choice. 
There is no role for the government or other OSH-promoting agency 
in this hypothetical world. The worker will make a well-informed 
choice -among his opportunities, trading levels of safety and health 
associated with alternative positions against their wages. 

The Employer's Choice : Wage Costs Versus OSH Expenditures 

Will the worker's choice lead to efficiency for society as a whole? 
Yes, if employers also follow the dictates of the competitive model. 
The wage premiums that must be paid to workers accepting health- 
threatening employment are the instruments that guide them to ap- 
propriate decisions. Employers weigh the financial costs they incur 
when they impose signilicant safety or health risks on their workers 
against the costs of reducing those risks. The optimal level for such 
risks in any workplace is set where the savings in wages from provid- 
ing a more healthful workplace would just not cover the additional 
costs of achieving it. (Thus it is the same level that the workers them- 
selves would set if they owned the enterprise, leaving aside the fact 
that their trade-off for safety and wages might change with newfound 
wealth.) This is the optimal level of safety, for it minimizes the sum 
of the two classes of costs we would like to avoid : expected accident 
costs plus the costs of preventing accidents. 

The competitive market achieves its efficient outcomes by bartering 
wage premiums against incurred risks.^ Thus, at the competitive equi- 
librium we would expect to find that among individuals who are other- 
wise alike, those who assume greater risks will be receiving higher 
incomes. The economics of markets for risky forms of employment 
devotes great attention to the existence and magnitudes of these wage 

Empirical Evidence 

Although recognition in the economics literature of the role of wage 
premiums for hazardous work dates back at least to the 18th century 
writings of Adam Smith,^ empirical studies to measure the actual 
magnitude of those premiums have appeared only recently. The most 
obvious manifestation of such premiums is in the hazard pay provi- 
sions of many contracts. A study by the Bureau of Labor Statistics 
(BLS) of 503 major collective bargaining agreements, both before 
and after enactment of the OSHAct, found that about 29 percent re- 
referred to such hazard pay.^ Most hazard-pay clauses were found in 
the construction industry, and the most common danger referred to was 
that of falling. Where the contracts set a specific amount of compensa- 
tion for hazardous duty, the amount specified most often was 25 cents 
per hour.^ 

3 Some of these premiums may be in a form other than higher monetary wages, such 
as pleasant working conditions or more extensive health and disability benefits. 

*A, Smith, The Wealth of Nations (Modern Library. New York, 1937). 

5 U.S. Bureau of Labor Statistics, "Major Collective Bargaining Agreements in Selected 
Industries : Safety and Health Provisions Before and After the Occupational Safety and 
Health Act of 1970." Report prepared by the Office of Wages and Industrinl Relations 
for the Office of Occupational Health Statistics, Bureau of Labor Statistics, Washington, 
D.C., February 1975. 

"Ibid., ch. VIII. 


In most cases, the additional wages paid for hazardous work are 
not explicitly identified in collective bargaining agreements. Walter 
Oi cites an unpublished study by K. Gordon, who found that wage 
differentials across U.S. Class I railroads were approximately equal to 
the differences in lost w^ages due to injuries.' Richard Thaler and 
Sherwin Rosen used insurance company data on the risk of mortality 
in various occupations to calculate the '"value of saving a life." Using 
multiple regression, a technique that holds other factors equal when 
estimating the effect of a particular variable, they found premiums 
for risks to life which implied a valuation of between $176,000 and 
$260,000 per expected life lost (i.e., a premium between $176 and $:^60 
for exposure to a .001 risk of death). An alternative (log-linear) 
specification of tlieir model yielded somewhat lower estimates, with a 
range of $186,000 to $189,000 using the mean values. Their estimates 
suggest that unionized workers receive higher premiums for risky 

Robert Smith also employed regression teclmiques to estimate wage 
premiums, using data tabulated by industry rather than by occupa- 
tion. He found that the premiums for non-fatal injuries were insig- 
nificant, but that the premiums for the risk of death were very large^ 
approximately $1.5 to $2.6 million per expected life lost.^ In contrast, 
James Chelius, who used data on individual firms in thirteen states, 
found (with a four ecjuation simultaneous equation model) no evi- 
dence of wage premiums in higher-risk industries.^^ 

In the studies cited above, the assumption is made, at least im- 
plicitly, that workers are awaie of the risks that they face in different 
industries or occupations. W. Kip Viscusi has approached the problem 
of estimating the impact of death and injury rates somewhat diff'cr- 
ently, employing survey data on workers' ovrn perceptions of the 
riskmcss of their jobs. Using multiple regression techniques to control 
for worker and job characteristics, he found that workers who re- 
ported that their jobs exposed them to dangerous or unhealthy con- 
ditions received on average about $375 more per year than those who 
did not. It should be noted that over one-half (52.2 percent) of the 
blue-collar workers surveyed claimed to be exposed to dangerous or 
unhealthy conditions. Viscusi also estimated the same equations using 
the BLS injury rate instead of workers' perceptions of their risks. He 
found that workers exposed to the average manufacturing injury rate 
received about $420 annually for incurring that risk.^^ Viscusi's results, 
consonant with those of Thaler and Rosen, suggest that workers in 
risky occupations are able to command higher wage premiums for 
risks if they are unionized. 

Viscusi also shows that quit rates are substantially higher for haz- 
ardous employment. ^2 Although alternative explanations relating to 
changing preferences or skill levels are possible, these differential quit 
rates seem to suggest that workers learn about hazards on the job. To 

7 W. Y. Oi. "On the Economics of Industrial Safety," p. 693. 

* R. Thaler and S. Rosen, "The Value of Saving a Life." pnrer presented at NBER 
Conference on Income and Wealth, Household Production and Consumption, Washington, 
D.C., November 30. 1073. 

»R. Smith. «a'he OSHAct. Appendix B. 

^" J. R. Chelius, "The Control of Industrial Accidents : Economic Theory and Empirical 
Evidence." law and Contemporary Problems. MS (1074), 700-729. 

*i W. K. Viscusi, "Employment Hazards : An Investigation of Market Performance," 
Ph. D. dissertation, Harvard University, 1976, ch. 12. 

" Ibid., ch. 11. 

174 . 

the extent tliat oovernmeiit intervention might provide information 
that would improve initial perceptions, or change risks so that a large 
number of Avorkers would not l>e drawn initially into hazardous em- 
ployment, it might prove beneficial, particularly if worker mobility 

The 'studies discussed above suggest that as economic theory pre- 
dicts, market forces do lead to higher wages for workers in risky pro- 
fessions. This implies that workers are at least to some extent cognizant 
of the occupational risks they run. It also implies that if such risks 
Avere to be eliminated, say through government regulatory efforts, 
those workers would likely suffer a loss of income. What these studies 
do not tell us is how accurately workers assess the heightened risks 
they run or how fully they are compensated. The studies are aggrega- 
tive and may not distinguish, for example, among workers at widely 
varying risks within the same occupation or industry. The studies 
also do not tell us how workers get sorted into different occupations, 
whether, for instance, workers presently in safe occupations would 
demand a rouo-hly equal or much more substantial premium to incur 
higher risks. Information on this issue might affect the implications 
for equity of the finding that wage premiums accompany hazardous 

In sum, the mere existence of wage premiums for risk does not indi- 
cate whether the market is working effectively. It suggests that the 
labor market recognizes risks, at least partially. Substantially more 
elaborate studios than those conducted to date would be required to 
determine whether such risks were fully recognized. 

If such risks were fully recognized by both employers and workers, 
and if preferences and institutional structures were such that a 
worker's health status was only his own concern, we could be assured 
that the outcome of competitive market processes would be efficient. 
Any Government intervention to promote or alter levels of OSH 
would by necessity reduce the welfare of some parties.^* 

We do not attempt in this analysis to assess the overall performance 
of labor markets, an analysis that might lead us to a general prescrip- 
tion for government intervention or nonintervention. Rather, we shall 
attempt to identify areas where the justifications for some govern- 
ment OSH-promoting efforts are weakest or strongest, and propose 
that resources be reallocated toward the latter. We shall also be con- 
cerned with identifying the t^^pes of government programs that seem 
most likely to generate efficient outcomes. 


The language establishing OSHA defines its primar^^ objective : "to 
assure so far as possible . . . safe and healthful working condi- 

" This argument, however, should be distinguished from a similar argument that market 
outcomes will not be efficient : The market will not provide premium pay for exposure to 
unknown hazards. One example cited is vinyl chloride, which has been s*^hown to increase 
the chances of developing angiosarcoma, a rare type of cancer of the liver. In that case 
workers were exposed to the hazard for several decades before the risks became known. 
The fact that vinyl chloride workers were not compensated for the extra risks they faced 
does not mean that the market operated inefficiently. Given the information available, 
market forces probably allocated resources efficiently. Until the risk was recognized, gov- 
ernment regulation was also impossible. 

" Even if labor markets do satisfy the competitive conditions, the government may 
have a role as a generator of new Information about the relationship between potential 
hazards and OSH. 


tions." ^'' Thorp are at least two Hasso? of difficiiltios with this charge. 
First, in conception it leaves aside other valued objectives, the most 
important of whicli is that appropriate value be attached to the re- 
sources of society directed to OSH promotion. Second, it lacks the de- 
gree of precision needed to make it applicable to and employable in 
policy formulation. No quantitative interpretation is given to levels of 
risk that might be regarded as safe or healtliful. 

How might appropriate levels of risk be defined? Zero risk levels 
are clearly ina])]n-o])riate. In many circumstances they are technologic- 
ally infeasible. In othei's. er'onomic considerations make it clearly un- 
desirable to eliminate all OSIi risks. Some appropriate nonzero level 
of T-isk must })e defined. This suggests that OSII A will ]>e confronted 
with tlie need to make trade-offs, implicitlv or explicitly, between 
OSH levels and the resources (including OSIIA's resources) required 
to achieve them. This is precisely the type of trade-off that is made 
implicitly when we choose not to further lower speed limits, when we 
impose limits on expenditures for medical care, or fail to require fire 
escapes on single-family dwellings. A critical question for policy is 
Avho should be making the trade-otfs in the OSH area. 


A reasonable response to that question is that the woiker himself 
should be the judge. He is the one whose health is Ix^ing threatened. 
If the worker is the judge, the trade-oif rate will be his vrillingness-to- 
pay for increased OSH levels. Moreover, if labor markets were perfect 
and competitive, wage premiums accompanying risk would just reflect 
willingness-to-pay. If laboi- markets are not perfect, in which case 
there is a rationale for government intervention, benefit-cost analyses 
that value OSII losses at workers' willingness-to-pay will lead to effi- 
cient outcomes. 

Xo issue in OSH has generated more vitriolic disagreement than the 
question of the appi()]n'iate weighing of economic costs. The argu- 
ments against the benefit-cost approach include : This is not what Con- 
gress intended when it made OSII a matter of public policy. Life is 
sacred and should not be bartered for dollars. Health and bodily integ- 
rity are not traded on markets: the quasi-market measure willingne^s- 
to-pay is therefore not an at)[)ropriate indicator of value. (Even if 
Avillingness-to-pay were acce])ted, there would be substantial disagree- 
ment on issues of calibration from sui'veys or market studies, whose 
willingness-to-pay, before or after injury, etc.^*^) 

Fortunately, substantial t)rogress in the reform of OSHA's proce- 
dures can be achieved without attempting to resolve the problem of 
valuing lost lives or impaired health. OSHA's procedures liave been 
formulated with minimal concern for economic resources. It should 
not be surprising, therefore, that there are substantial inconsistencies 
across areas in the cost of achieving OSH improvements. For a particu- 
lar firm, compliance with either standard A or standard B may reduce 
the number of expected accidents by .1 per year, "i'et compliance with A 
may cost $1,000 per j-ear whereas compliance with B may cost 10 times 
that amount. We could achieve more OSH for the same workers for 

^••OSHAcr. sect. 2(1)). 

^'' For a discussion of various methods for valuelnfr llfesaving interventiona, see R. Zeck- 
hauser. "Procedures for Valuing Lives," PuOlic Policy, 23 (1975), 419-464. 


the same resource expenditures if we enforced standard A more vigor- 
ously and abandoned standard B. All parties could be made better off 
if regulatory efforts were adjusted so that low incremental cost means 
of increasing OSH levels were pursued more vigorously and hi^h 
cost measures relaxed, so that the marginal costs of promoting OSH 
were made constant. 

This argument is less compelling, although only slightly so, if we 
look across industries or classes of workers. Most would agi'ee that we 
should not spend $5 million to save one life, say through safety rail- 
ings, while sacrificing another, perhaps through exposure to cotton 
dust, to save $50,000. We should formulate a regulatory system that 
attempts to achieve the greatest OSH gain for the levels of expendi- 
ture it entails, what economists would call a cost-effective system. 

Significant gains could be achieved if OSH-promoting efforts were 
rationalized so that resources were directed where they were most pro- 
ductive. In the hypothetical example above, for instance, 99 additional 
lives would be saved if $5 million were redirected from safety railing 
expenditures to efforts to reduce cotton dust exposure. If this process 
of resource shifting continued, always seeking the area where the 
greatest OSH gain could be achieved for a given level of expense, a 
cost-effective outcome would be reached. One of its significant char- 
acteristics would be that the cost of securing additional OSH would be 
the same in all areas of expenditure. This condition, familiar from 
economics, is that for an efficient outcome the marginal cost of achiev- 
ing a particular benefit must be everywhere the same. 

We conjecture that the gains in OSH would be much greater if we 
were to rationalize our present expenditures to be cost-effective than 
if we were to increase all present areas of expenditure by say 10 per- 
cent. The pursuit of a cost-effective outcome would yield an additional 
benefit: It would automatically generate information on how much 
additional increases in the level of OSH would cost. This information, 
presently unavailable, would be made accessible to the political process. 
Congress and the executive branch would have relevant information 
for decisionmaking. A major policy conclusion of this analysis is that 
the interventions of OSHA should be rationalized so that they follow 
the principle of cost-effectiveness. 


Three classes of observations introduce equity considerations into 
the discussion of appropriate policy for OSH: (1) health and safety 
threats vary in a recognizable manner from occupation to occupation 
and industry to industry; (2) those who suffer the highest OSH risks 
are differentially those with low earning opportunities;^^ (3) those 
who suffer serious losses of health, though perhaps receiving lost wages 
and medical expenses, are rarely compensated sufficiently to make 
them whole. Moreover, some forms of OSH-promoting interventions 
inhibit market forces which generate higher wage levels as accompani- 
ments to OSH risks. 

Proponents of government programs for OSH frequently express 
the belief that every American worker has the right to a safe and 

i''" This does not Imply that wage premiums are not being paid for assuming risks, since 
Individuals differ substantiallly in marketable labor skills. High skill people, in most 
cases, are "paid" in both higher OSH levels and higher wages. 


liealthfiil workplace.'® They may argue further that it is an inalienable 
right, one not for sale. Xo humane society should olfer emplo^anent 
that imposes significant risks on its workers' health. Individuals pres- 
ently working m high risk industries, they might observe, have little 
opportunity to switch. Indeed, it is increasingly difficult now to at- 
tract people to industries, such as coal mining and asbestos, that are 
known to be risky. The kernel of this argument is tliat workers, par- 
ticularly those disadvantaged in other ways, should not be forced to 
run significant health risks merely to earn a minimal standard of 

Those who support market outcomes would point out in response 
that premiums are being received for the risks that are being run. 
Moreover, the government, by prohibiting certain classes of activities, 
might end up hurting those it intends to help. (Which groups will be 
hurt and which will benefit in any case will depend on supply and 
demand elasticities for labor.) Why has this argument failed to con- 
vince the proponents who view safe and healtliful workplaces as a 
right? First, there is the special nature of health, a nonmarket good 
provided at least initially by nature. Second, discomfort arises from 
observing the correlation between occupational risk and the distribu- 
tion of income. Losses of life and limb draw attention to income in- 
equalities that would otherwise be fuzzier in our consciousness.'^ But 
the better remedy is to provide more income to poor people, not to 
deny them opportunities. 

It might help to get our thinking straight on some of these issues 
if we considered some hypothetical questions. What would happen if 
we banned all hazardous employment? Would our feelings about 
market determined OSH conditions differ if we had more substantial 
programs to transfer income ? Given such programs, would risks di- 
minish f AVould equity become a less compelling argument in relation 
to OSHA and OSH promotion ? Many areas of social policy are exceed- 
ingly inefficient places to bring about the redistribution of income or, 
more generally since we are including health in our discussion, of wel- 
fare. Occupational safety and health seems to be one of them, particu- 
larly if, as some suspect, some of its efforts work to the disadvantage 
of low-opportunity individuals, those who may welcome additional 
intome even at the expense of additional risk. 

The critical question for the equity discussion is : If workers w^ish 
to make a well-informed choice to assume a higher level of risk in re- 
turn for greater wages, and if there are no externalities, does society 
have tiie right or obligation to interfere with that decision? If the 
answer is no — and ours is — the govermnent's appropriate interven- 
tionary role to promote occupational health and safety is better de- 
lined. The government should look for situations where the market is, 
or is likely to be, performing inadequately, and where regulatory in- 
tervention can be expected to improve matters. 

^^ For example, Rep. Phillip Burton, a member of the House Education and Labor Com- 
mirtf-e wiien it reported on the OSH Act, stated in his "Separate and Concurring Views" : 
•'This bill represents a long-overdue significant recognition that working men and women 
need I>deral assistance to secure their inalienable right to earn their living free from the 
ravages of job-caused death, disease and injury." [italic supplied], as quoted in Bureau 
of National AfTairs. The Jol) Safety and Health Act of 1970 (Bureau of National Affairs, 
Washington. D.C. 1971). p. 199. 

^^ For a fuller discussion of these issues, see Zeckhauser, "Procedures for Valuing Lives." 



The issues of equity discussed above arise even when the competi- 
tive market functions perfectly, that is, efficiently. However, the ef- 
ficiency of market outcomes is called into question by several imperfec- 
tions. Our discussion focusses on three issues: (1) imperfect informa- 
tion, (2) institutional rigidities, and (3) externalities. 

Imperfect information 

Risks to health and safety are difficult to estimate. Given the im- 
portance of the area, we might expect that there would be extensive 
data series relating occupational conditions to health and safety con- 
ditions. There are not. No individual would have the incentive to 
tabulate these data himself, for he would have to bear all of the costs, 
but would reap only a small portion of the benefits.-*^ It would not be 
efficient for a private business to do so; to recoup its costs, it would 
have to charge recipients more than the marginal costs of providing 
the service. Moreover, it would have to make resource-wasting, though 
profits-conserving, eli'orts to pre\'ent noncustomers from securing this 
information. In some instances a national union which represents the 
vast majority of workers concerned with particular 0811 findings 
might allocate an efficient level of resources to generate information 
on OSH.^^ In general, however, given the public good aspects of OSH 
information, we must expect the Federal Government to support 
efforts to collect information on OSH, both statistical information 
from the field and experimental information from the laboratory. 

Some forms of OSH-related statistical information would not be 
worth collecting even if the benefits to all citizens were taken into 
account. Conceivably, then, the government has been efficient by not 
seeking to provide useful information on the relation between OSH 
experience and various hazards. The evidence suggests, however, that 
the predominant reason is that no division within the government has 
taken as its responsibility the need to generate both a consistent data 
base and studies drawing upon it that could infoiTn worker, employer, 
and government decisions." The Federal Government can play a 
useful role by gathering and processing information on OSH risks, 
and making its results known to all at a charge no higher than the cost 
of distribution. 

Even if relevant information on OSH were provided, workers 
might have a difficult time making informed decisions. Experimental 
and empirical evidence suggests that individuals make poor decisions 
where low probability events are concerned.^^ In their decisionmaking 
they may find it difficult to distinguish operationally between a .0001 
probability and a .00001 j^robability. There may be enough poor 
assessors or processors of probabilities to fill the riskiest professions 

-0 W. K. Viscusi's research suggests that workers do draw inferences based on the acci- 
dent experiences of their fellow workers. 

^ In 1970, the United Rubber Workers negotiated contracts specifying that the firms 
involved would contribute $.005 per worker per hour to a fund for research on the OSH 
risks facing n)embers of the union. 

2- There is the danger, of course, that once such information is generated there will 
be irresistible pressure for the government to undertake poorly considered interventions. 

^ For experimental evidence, see A. Tversky and D. Kahneman, "Judgement Under 
Uncertainty: Heuristics and Biases," Science, 185 (1974), 1124-1131. For an analysis 
of survey data on decisions to purchase flood and earthquake insurance, see H. Kunreuther,. 
"Limited Knowledge and Insurance Protection," Public Policy, 24 (1976), 227-262. 


with those who by chance underestimate OSH risks.^* (The inference 
from this line of reasoning is that if workers miderstood the risks 
they faced they woukl not assume them for the wage premiums now 

OSHA standards can be seen as a device for disseminating informa- 
tion. In that case, though, the Government has gone one step further 
than gathering information on the magnitudes of risks associated 
with various conditions. It has. in theory at least, processed the infor- 
mation and chosen the optimal response, which it identities as a stand- 
ard. The potential difficulty with this approach is that the Govern- 
ment will not have available infonnation pertinent to particular 
workplaces. A standard can not recognize lirm-to-tirm ditferences in 
the costs of meeting a standard, or in risks associated with particular 
conditions — for example, skill level of workers or frequency with 
which a machine is used — or differences among workers in their will- 
ingness to accept particular risks in return for wage premiums. 

As difficult as infonnation on risks of accidents may be for an indi- 
vidual to gather and process, the information on health risks is many 
times less tractable. Losses in health may not show up for many years. 
Moreover, precise attribution of an incident of health loss to employ- 
ment may be exceedingly difficult if the disease can have other causal 
factors. For example, even the most observant workers would prol)- 
ably fail to notice that working with substance X raises the risk of 
lung cancer 20 years later by 2 percent. 

Institutiayial rigidities 

Even if employers and employees have the appropriate informa- 
tion on safety and heaUh risks, various factois may pievent them 
from responding as they would weie OSH and employment decisions 
made in a competitive market. Where employment decisions are con- 
centrated in the hands of large unions and large employei-s. some 
forces that would otherwise push toward optimal health and safety 
levels will be discouraged. If wages are to be set on an industry-wide 
basis, a single employer will have a diminished incentive to attempt 
to provide a safer or more healthful workplace; his wage bill will 
not be affected. (Any interference with market processes for wage 
determination, foi- example, minimum Avnge legislation, may also 
have this effert.) The result may be to make health and safety condi- 
tions as well as wages more of a matter for bargaining. lender such 
circumstances, the union finds itself in an adversary relationship with 
the employer with regard to the determination of safety levels. If 
only one element were being bargained, say the wage, we could be 
relatively confident that an efficient outcome would be achieved. With 
many issues on the table, it is quite possible that the ultimate agreed 
upon packajie will be inefficient, in the sense that both the employer 
and the employee would prefer an alternative package of conditions.-'^ 

^ Xo bias in asspssnient is implied liore. There -vrill also be a ffronp of "over-estimators" 
which will choose safer employment than would be optimal for them. The overall dis- 
tribution of risks in employment will not be the same as If all individuals could predict 
risks accurately. 

-"' .\-- a tt>chni<'al ]«>int. it should be reco^nize<] that safetv levels for all workers in a firm 
are likely to be set simultaneously. Tliey will not be chosen individuallv on a worker-bv- 
worker basis as was implied in our discussion of the comT)etitive model. The safetv level in" a 
firm may be what econcnnists refer to as a public pood. There is the possibility that the 
level of provision will be inefficient, either too hiffh or too low. if the firm looks onlv to 
the preferences of the marginal worker when deciding whether to raise or lower its safetv 


'When the government imposes a minimum safety level, the nature 
of the bargaining process is changed. Whether outcomes will be more 
or less efficient is difficult to predict. 

A variety of institutional rigidities inhibit market forces that would 
convey financial reward to an employer who provides a safer or more 
healthful workplace. Insurance policies for workmen's compensation 
or employee health coverage are much less than fully experience rated 
due to the nature of the insurance industry and its regulatory struc- 
tures. That is, an employer's OSH record is not sufficiently taken into 
account when his premiums are set. Employers thus pay much less than 
the resources lost when their workers get sick or injured ; OSH suffers. 
The array of factors that limit worker mobility also works against 
OSH ; workers will not respond fully to differentials in OSH condi- 
tions. In particular, when a productive process or workplace is dis- 
covered to be unhealthf ul, workers may find it costly or difficult to re- 
move themselves. 

The presence of such rigidities implies that market processes will not 
achieve an efficient outcome. No one particular form of government 
intervention is indicated, though it does suggest that measures to limit 
institutional rigidities should be considered. 


The competitive market could only work efficiently if no one other 
than the employee and his employer were concerned with the level of 
health or safety provided him on his job. If the welfare of others is 
affected, then there will be externalities to his work choice. Competitive 
market processes will not be efficient for they will neglect these ex- 
ternalities. Two types of externalities may be present: physical and 
financial.-^ A physical externality might take the form of a worker 
contracting a communicable disease, or carrying a health-endangering 
substance outside the workplace on his clothes. Though evidence estab- 
lishes the existence of this form of externality in some cases (e.g. for 
asbestos, lead, and kepone), the occupational health hazard that con- 
veys a physical danger to others is clearly the exceptional case.-^ Physi- 
cal externalities thus cannot serve as a major justification for Govern- 
ment intervention in markets for occupational safety and health. 

Financial externalities are generated by the institutional arrange- 
ments of our society to care for and compensate individuals who are 
sick or injured, including those who suffer an occupational illness or 
disability. Life insurance benefits, health care costs, the expense of sup- 
porting a family whose wage earner is ill or deceased are borne only in 

2» Some analysts might identify what could be labelled an attitudlnal externality, related 
to the special role our society accords to health. Health is sometimes referred to as a 
merit good ; it is suggested that unlike clothes or televisions, society has an obligation 
to provide all of its citizens the resources required to achieve maximum available health. 
The existence of extensive government programs to subsidize health care shows the 
strength and persistence of this attitude. 

The authors of this report, feeling strongly that individuals' preferences should be 
respected wherever possible, question the wisdom of our government's efforts to redirect 
resources towards health that would otherwise be spent for other purposes. We would 
argue in particular that those receiving these resources would be better off if they could 
spend them as they chose. 

Should arguments supporting this attitudinal externality be persuasive, it would then 
be appropriate to measure the externality's magnitude. It may be far smaller than would 
be inferred from our present array of programs. 

^ The same processes causing hazards to worker health may, of course, also release 
pollutants into the general environment, thus threntf^ning the health of nearby residents. \ 
Such effects, however, are probably dealt with more appropriately by an agency concerned 
with environmental pollution, such as EPA, rather than by OSHA. 


small part by the worker whose health is impaired or by hi? employer. 
Government-provided insurance, including social security for dis- 
ability and welfare payments, is not experience rated. Employers 
who impose higher costs on the insurance fund by permitting un- 
healthful working conditions are charged no more. ^Moreover, our 
whole medical care system is laced with a web of governmental subsi- 
dies. "When a worker gets sick, the rest of society, directly or indirectly, 
foots a substantial portion of the bill through its taxes. Even private 
insurance schemes, say an employer's health benefits program, are far 
from fully experience rated. (Some insurance companies providing 
coverage for workmen's compensation inspect facilities as a means 
of avoiding undue risks, and in effect engage in merit rating.) Should 
some form of comprehensive health care coverage be enacted under a 
national health insurance plan, these financial externalities would be- 
come more direct and substantially magnified in scope. 

Given the existence of these externalities, neither an individual nor 
his employer will take into account the full costs associated with the 
risks to the employee of injury or health loss. Government intervention 
to promote OSH can be viewed as one policy measure attempting to 
secure an outcome that appropriately weights these externalities. If, 
for example, exposure to a particular occupational health risk imposed 
on society an expected future cost of $100. competitive processes could 
only lead to an efficient outcome if this $100 cost were cliarged to the 
labor contract. ^^ "What if the Government wished to employ not a tax 
but a standards system? In theory at least, it could predict the OSH 
conditions that would result from competitive processes if this ex- 
ternality tax were imposed. It might then specify those OSH condi- 
tions as a standard, using any of a variety of enforcement systems. (In 
praetice, tlie government wouhl be una})le to vary tlie standard across 
areas of application so as to recognize firm-to-firm variations in costs 
and OSH implications.) 

Xote here the phenomenon of pyramiding intervention. The govern- 
ment starts by subsidizing health care costs. This then reduces the 
incentives of individuals to pursue efficient levels for their own health. 
But society as a whole, throuijh its sul)sidy schemes, has secured a 
financial interest in individuals maintaining their health. This sug- 
gests that the government must intervene once airain. to assure that 
freely accepted health risks are not excessive from the standpoint 
of efficiency. ^^ 

When considering interventions to deal with externalities, it is not 
sufficient merely to identify tlie existence of an externality. Its mag- 
nitude should be estimated as well. It would seem, for example, that the 
magitude of externalities, measured either in absolute terms or relative 
to the costs to the individual and his employer, is generally greater 
where health losses, as opposed to safety losses, are concerned. Work- 
men's compensation imposes costs on the employer in the second in- 
stance; there is no equivalent charge for the long-term health losses 
his employment may generate. Among health losses also, we might 
wish to distinguish magnitudes of externalities. If the effect of ex- 

*^The nUiniate outcome would he the same whether the emplovpe or emnlovpr were 
char-pd. The elastic discussion on tlds suh.iprt is R. Coase. "The Problem of Social Cost," 
Journal of Law and Economics. 3 (19fiOK 1-44. 

**The advent of national health insurance may strengthen the incentive for the Gov- 
ernment to intervene in a variety of areas where health is endangered. 

83-944—78 13 


posure to noise is to induce hearing loss, the primary effect of which 
in turn is loss of personal pleasure, then the individual worker will 
be bearing most of the cost of this health loss. The asbestos worker, 
who is at high risk of cancer, is quite a different phenomenon. Society- 
loses from both having to support his medical treatment and possibly 
caring for his family after he is dead or no longer able to work. 

In sum, present, societal institutions generate financial externalities 
from workers' and employers' OSH decisions. The existence of these 
extrenalities argues for consideration of sensible government inter- 
vention to promote occupational safety and health. 



A number of inferences should be drawn from this discussion of 
health and safety determination. First., if all markets ^° were competi- 
tive and if information were perfect, an efficient level of OSH would 
be achieved. There would be no way to secure a different level of OSH 
without diminishing the Avelfare of some individuals. There is em- 
pirical evidence suggesting that the market recognizes occupational 
risks, though perhaps not full}^; this evidence is better documented 
for safety than for health.^^ Second, strong equity arguments appear 
to motivate much discussion of appropriate government roles in the 
promotion of OSH, though the implications of these arguments are 
rarely spelled out. Third, market imperfections may impinge on de- 
cisions that affect OSH. The most significant imperfections include 
poor and difficult-to-process information, and externalities. Collec- 
tively provided compensation programs, as well as traditional insur- 
ance arrangements, create an important, though frequently overlooked, 
class of financial externalities. 

The confluence of these factors suggests that some form of gov- 
ernment intervention should be considered for markets affecting OSH 
levels. The potential benefits and liabilities of each possible form of 
intervention should be assessed in light of these factors. 

Efficient Prevention — The objective of regulatory intervention in 
this area is quite simply to increase levels of safety and health in the 
workplace. Two elements contribute to the efficiency of this effort. 
First, appropriate levels of OSH should be achieved. Second, whatever 
the level of OSH, it should be secured with the smallest sacrifice in 
other resources. (These elements may be in conflict. Suppose regulatory 
scheme A leads to OSH level A*. Regulatory scheme B is less efficient 
and would require more resources to achieve A*, but it is also more 
flexible and can be used to achieve higher OSH levels that are deemed 
more appropriate.) 

When considering efficiency, the broadest possible definition should 
be taken of cost; it includes any sacrifice of resources or well-being 
within the society. The efficiency objective of OSH regulation should 
be to minimize the sum of the cost of safety and health loss plus the 
cost of eliminating safety and health loss.^^ 

30 There would have to be markets for externalities as well as for contingent claims 
(goods received contingent on some particular outcome). 

»i Viscusi, however, actuallj^ discovered a larger wage premium for "unhealthy" than 
"unsafe" conditions. 

32 This is the spirit of the central argument in the classic volume by G. Calabresi, The 
Cost of Accidents: A Legal and Economic Analysis (Yale University Press. New Haven 


Appropriate Distribution of Costs and ^.<^^^^P^f ^i^ion-Rc^^^^^^ 
interventions will differ in the distribution of ObH and ot me 
resource^^^^^^^ to promote it. A scheme which taxes injuries, for 

TxampTe wUl prove costly to both those who own and those who work 
irpiesenUy dangerous industries: the government will reap some 
r^vlTues AchS an equivalent level of OSH by relying on a sys- 
tem thft^r^^^^^^^^^^ Resources to workers who suffer health loss will at 
east prov[de"some compensation to the workers -^^ -f-^^^^^^^^ 
ards system which establishes the same level of 9^^^^^""^^! ^^f.^^T^^ 
all markets were perfectly competitive, would yield the worKCis the 
«vime exDected financial rewards as the compensation system. 

xLXs^r bution of OSH will likewise be affected by the mode of 
re-ulotory intervention. Standards systems, for example, may narrow 
variations in OSH levels more than will tax or f ^^P^^^^^^^^^ff;,^^ 

A number of factors will affect the way society feels about the dis- 
tribution of costs and benefits of OSH regulation In general, there 
is a strono- bias a<rainst systems which impose substantial economic 
dislocations. Identifiable circumstances where firms are put out ot 
business or workers out of jobs are often unacceptable, even if the 
original health-threatening circumstances in which they operated had 
onlv borderline legitimacy. , . , ^ , 

No one would propose that OSH regulation be employed as a major 
tool for accomplishing favorable redistribution of resources withm 
our society. However, any evaluation of a regulatory intervention 
should assess the extent to which it is consonant with the society's over- 
all redistributional objectives. A criticism that is sometimes leveled at 
various modes of protective legislation is that they impose costs on the 
poor to benefit the middle class: that is, the latter group would gladly 
bear the resource costs required to generate tlie gains in safety and 
health, wdiereas the poor would rather take the risks and have the 

The distribution of health may be less easy to ignore than the distri- 
bution of dollar resources: it may also generate stronger feelings of 
guilt. A particularly pernicious possibility is that we may employ a 
program that protects the health of the poor, at the expense of their 
financial well-being, to becloud inequities in our present income distri- 

Finally, there is ample evidence that significant numbers of members 
of our society believe strongly that health, and by inference OSH, 
should not be provided through the traditional market processes of 
our societ3\ Often this may be a paternalistic belief: Some people, it 
is argued, particularly poor people, will not know enough to choose 
safe and healthful emplo3^ment. Others believe that, unlike entertain- 
ment or attractive clothing, the health of the citizenry is a responsi- 
bility of the society as a whole. Beliefs such as these run deeply and 
are found widely. They will and should influence our decisions about 
OSH regulation, and the way its benefits and costs are distributed. 

Administrative Efficiency — The history of American regulation is a 
history of disappointment : Performance has rarely matched expecta- 
tions. The cruel lesson is that it is exceedingly difficult for government 
regulatory agencies to generate desirable behavior on the part of self- 
concerned external parties. Government efforts to enhance OSH may 
encounter difficulties determining what conditions in the workplace are 


critical for improving liealtli and safety. There will be a host of 
administrative problems getting firms to change conditions that are 
identified as threatening OSH. Quite significantly, a system that 
requires government oversight will encounter difficulties monitoring 
conditions in the millions of workplaces in America. Any difficulties 
that OSHA may encounter in conducting its assigned functions will 
be compounded by the Agency's expectable desire to create administra- 
tive mechanisms that will highlight positive aspects of its performance. 
Less evident, though more valuable, areas of accomplishment may be 
slighted; more subtle regulatory approaches may be abandoned for 
those that create a strong impression of doing the job. 

Two lessons should be drawn when considering a regulatory inter- 
vention: (1) It will probably perform below expectation and (2) 
careful prior consideration of potential administrative problems may 
help in the choice of alternative regulatory formats. 

Provision of Information 

As noted earlier, government can usefully respond to a known 
market imperfection by collecting and generating information on 
the causes of occupationally induced accidents and illnesses. The ar- 
gument for government provision of information is much stronger for 
health than for safety, particularly since extensive statistics collected 
over a long period of time may be required to identify possible rela- 
tionships between health and exposure. When the government does 
generate and supply information, efficiency requires that no charge be 
made beyond the cost of dissemination. If merely generalized infoiTaa- 
ticn is being provided, say on an exposure-response relationship, a 
xero-charge system would seem desirable. However, if government 
experts are to come to a worksite and provide consultative services 
on the best means to meet a health or safety hazard, some charge 
might be assessed. Indeed, there is no reason why government and pri- 
vate consultants could not compete in the provision of such services. 

Incentive Approaches 

Workmen's Compensation — ^Workmen's compensation must be 
viewed as a major component of any governmental strategy for deal- 
ing with occupational safety and health. Compensation laws of some 
sort are in force in every State and at the Federal level.^^ The primary 
purpose of workmen's compensation is to provide workers with in- 
surance against lost earnings and medical expenses due to occupational 
injuries and illnesses. Transactions costs are low compared to those in- 
curred by pursuing compensation through court suits. 

In theory, at least, workmen's compensation has the added virtue of 
giving employers an additional incentive to provide safer and healthier 
working conditions. In practice, the incentive effects of workmen's 
compensation laws are greatly reduced by the fact that most firms in- 
sure, either with private companies or with State-run agencies. The 
very largest firms are fully experience rated or they self-insure ; that 
is, their workmen's compensation costs reflect in full their accident 
records. Small firms pay premiums based only on their industry and 
size classification ; their own records have no effect on their preniiums, 

33 For a review of State workmen's compensation laws and recommendations for their 
reform, see National Commission on State Workmen's Compensation Laws, Report of the 
National Commission on State Workmen's Compensation Laws (Washington, D.C., 1972). 


thus reducing incentives to provide safer conditions, mile it would be 
infeasible— and, from the standpoint of appropriate risk spreading, 
undesirable— to apply complete experience rating to small tirms, it 
would be possible to increase the incentive effects of workmen s com- 
pensation bv demanding that policies contain deductible and coinsur- 
ance rates tied to the size of the firm.^* In other words, firms would pay 
the full costs of compensation up to some limit, and then a declining 
proportion thereafter, with the remainder covered by insurance. ^ 

Worlvmen's compensation has the potential to provide firms with 
strong incentives to provide safer workplaces if changes are made to 
tie costs more closely to experience. Its potential for reducing health 
hazards is considerably smaller. Despite recent changes m the laws 
of many States, it is widely agreed that the overwhelming majority 
of victims of occupational illness receive no workmen's compensation 
payments.35 j^ js usually difficult, if not impossible, to determine 
whether a particular illness is of occupational origin, and. even if that 
determination is made, to attribute it to a particular period of employ- 
ment. For example, even with black lung, which is recognized as an 
occupational disease of coal miners, it may be impossible to decide 
which firm should be charged if the miner involved has worked in 
several different mines. 

Quite apart from any impact on incentives, the compensatory effects 
of workmen's compensation are desirable. A systematic effort should 
be made to insure that workmen's compensation levels correspond 
roughly to efficient levels, as might be judged by the levels workers 
themselves would choose if they were participants in some model in- 
surance scheme, where they would have to trade off their own wages 
against their own compensation amounts in an actuarially fair manner. 
In 1975, maximum weekly compensation levels ranged from about $60 
to more than $-250, with most States clustering in the range between 
$70 and $120. Only four States had monetary or time limitations on 
medical care provided for injuries. ^^ This is not the place to conjecture 
whether these levels are, in fact, efficient. The appropriate level of 
compensation will reflect a balancing of several objectives, including 
coverage of medical expenses, replacement of lost income, and pro- 
vision of efficient incentives to both workers and firms. 

Tort Law — Workmen's compensation is quite clearly intended to 
avoid the entanglements of tort law in providing swift and equitable 
compensation to injured workers. Under the master-serA^ant common 
law, the employer has certain legal duties with regard to his workers, 
namely : 

1. The duty to provide a safe place to work. 

2. The duty to provide safe appliances, tools and equipment for the work. 

3. The duty to gjive warning: of dangers of which the employee might reasonably 
be expected to remain in ignorance. 

4. The duty to provide a sufficient number of suitable fellow workers. 

5. The duty to promulgate and enforce rules for the conduct of employees 
which would make tlie work safe." 

3* This conclusion is also reached by L. Russell, "Safety Incentives of Workmen's Com- 
persntion Insiirance," Journal of Human Resources (Summer 1974), 361-375. 

^ NCSWCL Report 

3« F. C. Johnson, "Worker's Compensation Changes in 1974," Monthly Labor Review 98 
(January 197o), p. 32. 

s" Prosser. W. L.. Handbook of the Law of Torts, 4th ed., (St. Paul, Minnesota: West 
Publishing Co., 1971), p, 526. 


These sound very much like the responsibilities that motivate the 
programs of OSHA. The threat of liability under tort law might serv^e 
to provide employers with proper incentives with regard to OSH. The 
existence of compulsory workmen's compensation, however, rules out 
the general use of tort law to this purpose. Prosser states : 

When nn injury to a servant is found to be covered by a workmen's compensa- 
tion act, it is uniformly held that the statutory compensation is the sole remedy, 
and that any recovery at common law is Mrred.^ (Emphasis added.) 

The principle involved here is that the worker is assumed to trade 
the surety of a lesser settlement (statutory compensation) for the un- 
certain chance of a larger jury settlement at common law. Tims, 
wherever there is a compulsory worker's compensation law covering a 
worker, the statutory compensation is his exclusive recourse aga.inst 
his employer. The employer is thus immune from any action, provided 
he pays his insurance premiums. (There are a variety of possibilities 
for third-party suits, say against an equipment manufacturer, and 
complicating factors in those situations vv^here workmen's compensa- 
tion does not apply.) 

The superseding of tort law by workmen's compensation has been 
acceptable in part because when dealing with OSH problems tort law 
runs into significant difficulties. For many industrial accidents, indeed 
exposure to health-threatening conditions, it may be difficult for the 
courts to determine who is at fault. Courts are ill equipped to disen- 
tangle situations of partial responsibility, or to assess damages when 
only probabilistic judgments can be made about causality, as our ex- 
perience with the much simpler case of automobile accidents makes 
clear. Medical malpractice probably bears a closer relationship to lia- 
bility for OSH losses. Numerous workmen's compensation claims in- 
volve relatively small amounts; transactions costs would strongly 
interfere with the operation of the system. It is difficult to predict the 
financial outcomes of tort actions; the random element in compensa- 
tion would contribute to general citizen apprehension about and dis- 
taste for legal mechanisms. ^^ 

Taxes — Tort law and workmen's compensation provide firms with 
safety and, to a much lesser extent, health incentives by holding them 
accountable for the costs borne by workers. Tax schemes, in contrast, 
would transfer funds from firms to the government, the goal being 
to force firms to internalize the costs imposed on society as a whole 
when a worker is injured or ill. That is, taxes are intended to correct 
the distorted incentives created by externalities. 

The principle of levying taxes on offending parties when external- 
ities are present is well established in economics, particularly in rela- 
tion to pollution issues, where the taxes are referred to as effluent 
charges.^*^ The primary advantage of taxes over a system of uniform 
standards is that they allow firms to find the most efficient means of 
reducing the externality, and lead to an efficient outcome in the sense 
that for any given level of control, expenditures are minimized.^^ In 

28Ibid., p. 531. 

s» Movement away from tort law remedies is seen in the spread of no-fault automobile 
insurance plans and various proposals put forth to handle medical malpractice claims. 

*o The literature on the use of taxes to correct externalities is voluminous. For a non- 
technical exposition on the use of taxes to control pollution, see L. Ruff, "The Economic 
Common Sense of Pollution," Public Interest (Spring, 1970), 69-85. 

*i W. J. Baumol and W. E. Oates, "The Use of Standards and Prices for Protection of 
the Environment," in P. Bohm and A. V. Kueese (eds.), The Economics of Environment 
(London, 1971, 53-65. 


theory, standards could be set on a firm-by-firm basis to reflect varying 
costs and benefits; in practice, such a scheme would be unworkable. 
Furthermore, a tax system has the advantage that firms which do not 
or can not respond to the incentive pay a penalty. Thus even if the 
injury rate in a hazardous industrv^ is not reduced, the prices of the 
goods produced will rise, thus shiJrting demand towards goods pro- 
duced by less hazardous technologies. 

Robert Smith has proposed that an injury tax be substituted for 
safety standards.^- Such a tax, he argues, would provide more gen- 
eralized incentives for firms to improve safety conditions, stimulating 
them to attempt to influence the whole range of factors which con- 
tribute to accidents, and not just to control the limited number of 
physical conditions susceptible to direct regulation. Smith lias esti- 
mated, on the basis of an econometric model, that a tax of $500 per 
disabling injury would lower the injury rate by between 2.2 and 3.2 
percent, with more substantial reductions possible if higher taxes 
were imposed. For example, he estimates a tax of $2,000 would lower 
injuries by 8.8 to 12.5 percent.*^ 

An injury tax would be relatively easy to administer. No longer 
would it be necessary to inspect individual workplaces on a regular 
basis. Firms could either self-report, in the same way that they do for 
income taxes, or the tax system could be tied to workmen's compensa- 
tion claims. The latter approach would have several advantages. First, 
the administrative mechanism is already in place. Second, tying the 
fines to workmen's compensation would give workers an incentive to 
police compliance by their employers. Finally, to the extent that the 
level of externality is related to the level of compensation paid, the 
tax rate could be expressed as some fraction of the workmen's com- 
pensation claim. 

As Smith admits, a direct tax on occu])ational disease would be 
impractical given the difficulties of connecting individual cases of ill- 
ness with particular firms. It would be possible, however, to levy 
taxes on health-threatening conditions. This approach would be anal- 
ogous to the use of eiUuent charges for environmental pollutants. 
Such a system would be difficult to monitor, but then so is a system 
of performance standards for health hazards; both require that expo- 
sure levels be measured. 

The use of taxes and other incentive mechanisms to promote occupa- 
tional safety and health deserves more consideration and study. Their 
theoretical merits are well established, and their practical difficulties, 
while significant, may turn out to be no worse in most cases than those 
associated with standards. (For example, in order to maintain incen- 
tives it would be essential to work out a mechanism to prevent firms 
from insuring against the imposition of the tax.) One overwhelming 
obstacle to the adoption of OSH-related taxes we believe to be political 
in nature. Aside from economists, most people, including legislators, 
appear to lack faith in the power of this type of incentive system.^* 

*2R. S. Smith, "The Feasibility of an 'Injury Tax' Approach to Occupational Safety," 
Law and Contemporary Problems, 38 (1974), 730-744. 

^Mbid.. p. 741. 

<* A. Kneese and C. Schultze. Pollution. Prices, and Public Policy (Brookinp:s Institution, 
Washington, D.C., in7o), p. 116. sugj^est that one reason for "the congressional propen- 
sity to rely upon re?:ulation as the solution to complex social problems is the fact that 
most Congressmen are lawyers." 



OSHA has relied almost exclusively on direct regulation of safety 
and health in the workplace through the promulgation and enforce- 
ment of standards. Standards can be thought of both as a means of 
superseding the need for difficult-to-secure information as well as one 
means to deal with externalities (by imposing higher than market 
equilibrium levels of OSH). Most of the standards now in effect are 
"specification standards"; for example, they specify the height and 
construction of certain railings and the types of guards that must be 
mounted on particular types of equipment. A few are "performance 
standards" ; for example, they specify the maximum exposure to noise 
permitted, but give each firm fairly broad discretion as to the specific 
means by which it achieves that level. Standards of both types have 
their most natural application with regard to capital equipment, in- 
cluding both machinery and the plant itself. Standards covering 
training and supervisory practices are difficult to write, and still more 
difficult to enforce. Under the provisions of the OSHAct, OSHA does 
not have authority to impose fiLnes on workers for violations of 

Physical standards would seem potentially to be most effective when 
capital equipment is the prime determinant of accidents, and where 
unambiguous, easily monitorable standards can be imposed. Critics 
of OSHA frequently argue that the current regulatory effort is mis- 
guided, citing the widespread belief among safety professionals that 
worker behavior is such a critical element in safety. One of the earliest 
studies in the area, based on an examination of 75,000 accident case 
files by H. W. Heinrich, concluded that only 12 percent of all indus- 
trial accidents could be attributed to "dangerous conditions" ; the re- 
maining 88 percent resulted from "unsafe acts" by workers.*^ More 
recent estimates cited by the National Commission on State Work- 
men's Compensation Laws are less decisive and place greater emphasis 
on the interaction between worker behavior and the work environ- 
ment. W. Dean Keefer of the National Safety Council testified that 

18 percent of occupational accidents are due to environmental factors, 

19 percent to behavioral factors, and 63 percent to a combination of 
the two. The Pennsylvania Department of Labor and Industry re- 
ported that fully 95 percent of industrial accidents are due to a com- 
bination of human and mechanical problems, 3 percent to strictly 
mechanical failures, and 2 percent to human errors. Furniss estimates 
that 15.7 percent of the industrial accidents in Great Britain are 
caused by environmental factors, with the remainder due at least in 
part to worker behavior.^'' 

On a conceptual basis at least, it is impossible to say whether capital 
equipment regulations are more or less desirable, or should be more 
or less stringent, when other factors, such as worker care, are deter- 
minants of accident rates. Consider two hypothetical cases. In the 
first, situation A, workers are exceptionally cautious. Workers in the 
second, situation B, are careless. An anal^^sis of the two situations 

*5 The OSHAct, section o(b), directs that "Each employee shall comply with occupa- 
tional safety and health standards . . . which are applicable to his own actions and con- 
duct." No provisions are made, however, for imposing penalties on workers for noncom- 

*« National Commission on State Workmen's Compensation Laws, Compendium on 
Workmen's Comnensation (Washington, D.C., 1972), p. 287. 

*7Ibid., p. 288. 


would show that capital equipment was implicated in a higher pro- 
portion of accidents in situation A than in B. In which case vrill the 
imposition of safety requirements on machinery, say the installation 
of a guard, have a greater impact in reducing injuries? The intuitive 
answer might be, ''the guard will be most effective in situation A be- 
cause a higher proportion of accidents there are caused by capital 
equipment." Yet the guard might well do the most good in situation 
B, where it would prevent many workers from being injured as a 
result of their own carelessness. Of coui-se, if it were possible to pursue 
some intervention that encouraged workers in situation B to exer- 
cise more caution, that might prove to be a more cost-effective strategy 
than physical standards. 

More direct evidence on the potential efficacy of physical standards 
as a means of reducing occupational accidents is provided by studies 
from the States of California. Xew York, and Wisconsin. In a recent 
report to the Department of Labor, John Mendeloff discusses a study 
conducted by a panel of engineers in the California Division of Indus- 
trial Safety. Members of the panel reviewed a total of 920 injury 
reports. They concluded that only 18.4 percent of the injuries involved 
could have been prevented by a fully effective government safety 
inspection program. They estimated that an additional 8 percent could 
have been prevented by in-housc inspections.*^ 

A 196G review of a sample of 3,216 disabling injury reports filed 
under the Xew York safety consultants program yields results in close 
accordance with the California study just cited. Followup inspections 
there revealed that safety co<le violations were involved in only 22.4 
percent of the cases. However, in the 206 accidents judged serious, 57.3 
percent involved safety code violations.*^ 

Four studies from Wisconsin shed additional light on the ability of 
safety codes and inspection programs to reduce injur\^ rates. The 1971 
Wisconsin State Department of Labor. Industry, and Human Rela- 
tions effectiveness report estimated that 45 percent of the industrial 
accidents reviewed resulted from ''dysfunctional acts"' by workers, 30 
percent were due to transitory hazards, and only 25 percent could be 
attributed to identifiable physical hazards susceptible to inspection. 
The conclusion of the report cited this evidence in suggesting that 
worker training and education programs might be a better approach 
to reducing accidents.^^ Another Wisconsin study, which reviewed 
workmen's compensation claims, also suggests that violations of haz- 
ards covered by existing safety codes cause only a small proportion of 
accidents. Under Wisconsin State law. workers filing for compensation 
benefits receive an additional 15 percent if they can show that their 
disability resulted from violation of the State safety code. Of the 31,228 
claims settled in 1966, only 494, or 1.6 percent, included the extra 
payment. In recent years, the percentages have been still lower. The 
average payment to workers who successfully cited code violations in 

<* Oitert in J. Mendeloff. "An Evaluation of OSHA Propram's Effect on Workplace Injury 
Rates : Evidence from California through 1974." report prepared for the Office of the 
Assistant S^ecretary for Policy. Evaluation, and Research, U.S. Department of Labor, 
contract No. B-9M-5-2.399. Julv 1976. pp. 1.3-14. 

^ Cited in W. Y. Oi. "On Evaluating: the Effectiveness of the OSHA Inspection Pro- 
frr.ini." report submitted to U.S. Department of Labor, Contract No. L-72-86, May 15, 
1975. p. 42. 

" Cited in W. Y. 01, "On the Economics of Industrial Safety," p. 679. 


1966, however, was 2.72 times as high as the overall average.^^ This 
difference may suggest that code violations are more prevalent in seri- 
ous injury cases, or it may simply indicate that the difficulties of prov- 
ing a code violation outweigh the extra payment to workers in less 
serious cases, where the average payments are low. 

Two more recent Wisconsin studies of occupational injuries and 
deaths 'are reported by Nicholas Ashford. The first, on injuries, was 
based on a 10 percent sample of the "reported injuries investigated 
for causation by the Wisconsin Safety Specialists." The second set of 
data come from a study of all deaths investigated for causation by the 
Wisconsin Safety and Buildings Division. The studies concluded that 
30 percent of the injuries and 39 percent of the deaths resulted from 
violations of the Wisconsin Safety code. "Other unsafe conditions" 
were blamed for 24 percent of the injuries and 19 percent of the deaths. 
Worker behavior was implicated as the sole factor in 35 percent of 
the injuries and 26 percent of the deaths.^^ These data must be inter- 
preted cautiously, however, as they are based only on those accidents 
investgated for causation. No indication is given by Ashford of the 
criteria used to select cases for investigation. Common sense suggests 
that an injury or death would be more likely to be investigated if 
there were reason to suspect code violations. If that were the case 
estimates of the proportion of accidents caused by code violations based 
on these data would be biased upward. 

The evidence reviewed above suggests that even the most carefully 
drawn and rigorously enforced safety standards will not be successful 
in eliminating the majority of accidental injuries and deaths.^^ It 
seems unlikely that standards could eliminate even as much as a third 
of all accidents. A more realistic figure, one which takes into account 
the many difficulties involved in setting standards and the limited 
resources likely to be available to enforce them, would be far lower. 
This is not to argue that standards are necessarily an inappropriate 
instrument for promoting occupational safety; any reduction in in- 
juries is beneficial. But it does suggest that expectations for OSHA's 
performance should be modest. (Note, by contrast, that some form of 
incentive system would be appropriate for dealing with the broad 
range of causes of OSH loss. For example, management might be in- 
duced to promote safety training, change work rules, or appoint safety 
officers, none of which would be addressed by a standards system.) 

The discussion of the efficacy of standards thus far has been re- 
stricted to occupational safety. The authors of this report are not 
aware of any studies on the effectiveness of standards in reducing the 
incidence of occupational illness. In the health area, as opposed to its 
safety-oriented activities, OSHA has relied much more extensively 
on performance standards. For example, for new plants the coke oven 
standard specifies maximum permissible exposure levels, without re- 
quiring that particular methods of control be used. These perform- 
ance standards, while clearly desirable for the flexibility they allow 

^ Cited in W. Y. Oi, "On Evaluating the Effectiveness of the OSHA Inspection Pro- 
gram." pp. 43-44. 

82 Cited in N. A. Ashford, "Crisis in the Workplace" (MIT Press, Cambridge, Mass., 
1976), pp. 113-114. 

53 We have seen no evidence on the question : How do changes in physical equipment 
affect workers' tendencies to engage in OSH-threatening behavior? If an inspection system 
created an aura of safety consciousness, that would be a strong plus. 


firms in finding the least-cost means of control, may pose problems 
for monitoring and enforcement. The epidemiological studies on 
which most health standards must be based also usually provide only 
crude estimates of the impact of different exposure levels. It is inter- 
esting to note, however, that at least some of the most severe critics 
of OSHxi's use of standards in the safety area accept that approach 
to occupational health.^^ In part this attitude reflects the difficuhies 
cited earlier with regard to using incentive systems to promote occu- 
pational health. 

Standards and innovation 

Standards will tend to inhibit technological innovation, both gen- 
erally and more specifically with regard to safety and health promot- 
ing techniques. These problems are especially severe with specification 
standards. Firms have virtually no incentive to find new technologies 
for reducing risk if such technologies are likely to fall outside the 
narrow confines of the applicable standards. 

Even performance standards fail to create the appropriate incen- 
tives for innovation. Under performance standards, firms have an 
incentive to develop and implement cheaper alternative processes for 
meeting prescribed levels, but they have no incentives, beyond those 
already existing in the market, to develop technologies to reach safer 
or more healthful levels. In fact, if the strategy is to force firms to 
reduce hazards to the lowest level technologically feasible, there will 
be strong disincentives for conducting such research. Under an in- 
centive-based system — for example, experience-rated workmen's com- 
pensation coupled with an additional charge reflecting externalities — 
firms will have strong motivation to innovate since lower-cost tech- 
nologies will permit them to move to lower risk levels, in that way 
reducing expenses. 


Passage of the Occupational Safety and Health Act of 1970 was 
preceded by a long history of State and Federal Government involve- 
ment in the occupational health and safety area. Government regu- 
lated work practices and workplaces directly, and fostered systems 
to compensate those disabled by job-related injuries and illnesses. 

Common law Jidbility 

The earliest involvement of Government in occupational safety and 
health came through the courts. Under common law rules of liability- 
injured workers could sue their employers for negligence. The process, 
however, was slow and costly, and suits were difficult to win. In par- 
ticular,^ three common law defenses available to employers made the 
worker's chance of recoveiy extremely low: (1) The fellow servant 
rule, which permitted employers to escape liability if the worker's 
injury was due at least in part to a fellow worker's negligence: (2) 
the doctrine of contributory negligence, which denied recovery if the 
worker's injury resulted in part from his own negligence; and (3) 

" See, e.g., R. S. Smith, The OSHAct, esp. pp. 83-84. 


the theory of assumption of risk, which stated that in accepting em- 
ployment, the worker also accepted the risks commonly associated 
with his job.^^ As a result, the costs of occupational injuries and ill- 
nesses were borne almost entirely by workers and their families. 

Workmen^ s compensation 

TliQ original motivation for enacting workmen's compensation laws 
was twofold : (1) To provide workers with more equitable and reliable 
compensation for injuries than was available through tort law; and 
(2) to give employers stronger incentives to provide safer working 
conditions by forcing them to internalize a greater share of the costs 
associated with accidents. Workmen's compensation laws in the United 
States date back to the first decade of this century. By 1920, 40 States, 
the District of Columbia, and the territories of Alaska and Hawaii 
had enacted compensation laws of some sort, although most applied 
only to workers in particular, hazardous industries. By 1948, when 
Mississippi passed a workmen's compensation law, all States were 

Over the years, the coverage of workmen's compensation laws was 
extended, until in 1970, just before enactment of the OSHAct, approx- 
imately 85 percent of all employees were covered. In that same year, 
workmen's compensation cost emploj^ers about $4.9 billion, or 1.1 per- 
cent of covered payrolls." In most States, the amount of compensation 
to be paid for lost workdays is stated as a fraction of the worker's 
wage, most often two-thirds, with a limit on the maximum permitted.^® 
The National Commission on State Workmen's Compensation Laws 
(NCSWCL), formed under the OSHAct, found many deficiencies 
in existing laws. In its report, issued in 1972, the Commission noted 
that only 22 of the 50 States met even one-half of the 16 criteria 
suggested by the Department of Labor.^^ The Commission's recom- 
mendations included extension of coverage to all workers, increases 
in benefit levels, removal of dollar and time limits, and greater cov- 
erage of occupational diseases. 

The employer's incentive to provide a safer workplace under work- 
men's compensation is limited since most firms are insured. The 
NCSWCL estimated that only 20 percent of all firms (with 80 per- 
cent of covered payrolls) are large enough to be experience rated; 
that is, their premiums depend at least in part on their individual 
accident record and are not based solely on their size and industry 
classification.^^ However, only for the largest firms, those with more 
than 2,500 employees, is the experience rating complete; for smaller 
firms, the premium is a weighted average of the firm s individual 
record and the record of its risk class. Louise Eussell has examined 
the formula for rate setting established by the National Council on 
Compensation Insurance. She finds that even for a firm with 1,000 
employees, the elasticity of its premium payments with respect to its 
injury rate is only about 0.5; for every 1-percent increase in its acci- 
dent rate, its premiums increase hy only one-half of 1 percent ;^^ the 

55 Ashford, Crisis, p. 48. 

^ Ibid., p. 49. 

57 NCSWCL Compendium, p. 6. 

68 NCSWCL Report. 

69 Ibid., p. 119. 
60 Ibid., p. 96. 

81 Rnssell. "Safety Incentives of Workmen's Compensation." p. .?67. This elasticity is 
for a firm with an injury frequency rate of 0.05. Firms with higher rates experience a 
higher elasticity. 


responsiveness of premiums to injuries is substantially less for small 
and medium size firms. Kussell also finds that premiums are insensi- 
tive to the severity rate, a measure of lost workdays. She advocates 
a oreater use of deductibles and coinsurance as a means of increasing 
the ability of workmen's compensation to promote greater safety. 

Direct regulation of working conditions 

State efforts to directly regulate health and safety in the work-place 
date back to 1867, when Massachusetts formed a department of factory 
inspection. By the turn of the century, most of the highly industrial- 
ized States had passed at least some form of safety legislation.^^ gygn 
by the late 1960's, however. State efforts for the most part remained 
very limited. The Bureau of National Affairs reports : 

Some States had acted to establish and enforce safety and health standards, 
but only a few had modern laws and adequate resources for administration and 
enforcement. At least ei/2;ht States had no identifiable programs on occupational 
health at all. Four States had no inspection personnel and only 3 States had 
over 100 inspectors. Total State safety inspectors numbered only 1,GOO, which, 
combined with fewer than 100 Federal inspectors, totaled less than half the 
number of fish and game wardens in the country. 

Health and safety programs merited up to $2.70 per worker per year m some 
States, while others spent less than 1 cent.*" 

Regulation of occupational health hazards by the States was particu- 
larly deficient. A 1968 study by the Bureau of Occupational Safety 
and Health in the Department of Health, Education, and Welfare 
concluded that there was — 

No uniformity in the laws and rules and regulations issued to achieve this 
end, either in language or scope of legislation or in currency of standards, re- 
sulting in an inevitable disparity in effectiveness of protection afforded the 

Prior to passage of the Occupational Safety and Honltli Act. Federal 
regulation of occupational safety and health was limited to Federal 
employees, several particular industries, and firms holding Federal 
contracts.^^ The first Federal law in this area, passed in 1890, was di- 
rected at safety standards and inspection practices in coal mines. Over 
the next several decades, Federal safety regulation was extended to 
railroads; more important, a number of agencies concerned at least in 
part with occupational safety aiid health were formed, including the 
Public Health Service in 1902, the Bureau of ^Nlines in 1910. and the 
predecessor of the Occupational Safety and Health Division of DHEW 
in 1914. The AYalsh-Healey Public Contracts Act, passed in 1936, im- 
posed health and safety standards for workers employed under Gov- 
ernment contract. About that same time, the Bureau of Labor Stand- 
ards was created within the Department of Labor. In 1958, amend- 
ments to the Longshoremen's and Harbor Workers' Compensation Act 
imposed Federal safct}^ regulations on marine cargo handling. 

The 1960's saw a continued extension of Federal health and safety 
regulation to specific industries. The 1965 McXamara-O'ILara Public 
Services Contract xVct essentially extended the "Walsh-IIealey Act to 
suppliers of services to the Federal Government, and the IS'ational 
Foundation for the Arts and Humanities Act also included health and 

«2 > shford. Crisis, p. 47. 

«3BXA. .Tob Safety and Health Art of 1970, p. 14. 
8*Oi]nted in Aphford. Crisis, p. r>0. 

6^ This liistory of pro-OSHA Federal involvement is summarized from BXA, Job Safety 
and Health Act of 1970, pp. 51-52. 


safety provisions. Mining safety continued to receive special attention, 
with the Metal and Non-Metallic Mine Safety Act of 1966 authorizing 
inspections for certain health and safety hazards in the mining in- 
dustry. The Coal Mine Health and Safety Act, enacted in 1969, pro- 
vided for federally enforced health and safety standards, for aid to 
State and local programs, for research, and for compensation of miners 
disabled by pneumoconiasis, or "black lung." The Federal Government 
also became involved indirectly in the compensation of victims of occu- 
pational injuries and illnesses through the medicare and medicaid pro- 
grams, thus adding a new dimension to the financial externalities dis- 
cussed earlier. 


Legislative history 

Bills to establish general national occupational safety and health 
standards were introduced in the Congress as early as 1951.®*^ Over the 
next decade and a half, a number of similar bills were introduced, but 
none made much progress. Then, in 1968, President Lyndon Johnson 
proposed a program of research and development of comprehensive 
safety and health standards, the standards to be established and en- 
forced by the Secretary of Labor. Bills to accomplish that end failed 
to reach the floor of either House of Congress. The following year, in 
th© 1st session of the 91st Congress, occupational safety and health 
bills were introduced in both the House and the Senate, and the ad- 
ministration of President Kichard Nixon proposed its own bill. 

The major differences between the administration bill and those in- 
troduced by congressional Democrats included whether the power to 
promulgate and enforce standards should be lodged with the Labor 
Department, as the Democrats proposed, or with an independent Na- 
tional Occupational Safety and Health Board, as provided for in the 
administration bill. Additional disputes centered around such issues 
as a clause imposing a "general duty" on employers to provide a safe 
and healthful workplace, the right of employees to accompany inspec- 
tors, the conditions and procedures under which plants could be 
ordered shut down, and penalties for violations. 

Both the House and the Senate Labor subcommittees held exten- 
sive hearings on the proposed bills. Much of the testimony stressed the 
toll taken by occupational injuries and illnesses. It Avas estimated that 
each year about 14,000 workers were killed at work in accidents, and 
that 2.2 million suffered disabling injuries, resulting in financial losses 
of $1.5 billion in wages and $8 billion in the gross national product. 
Reliable data on occupational illnesses were much harder to obtain 
as it was widely agreed that only a small fraction of occupational ill- 
nesses were recorded in the official statistics. Extrapolations from one 
DHEW study indicated that 390,000 new cases of work-related ill- 
nesses occurred each year, with annual fatalities of 100,000. 

Evidence presented to Congress indicated not only the magnitude 
of the problem, but also its upward trend. Data on injury rates after 
the Second World War showed them falling through the mid-fifties, 

«« Unless otherwise noted, information on the legislative history is drawn from ibid. 
ch. 2. 


but then beginning to rise again. From 1960 to 1970, the injury fre- 
quency rate in manufacturing rose 26.7 percent, although the severity 
rate, a measure of days lost to injuries, held steady. Some, but not all, 
of the increase in the frequency rate could be attributed to C3xlical 
factors in the general economy and to the changing composition of 
the work force.^^ Data limitations made it impossible to determine 
trends in occupational illnesses, but testimony suggested that little 
was being done to limit workers' exposure to many known health 
hazards, and tliat new potentially harmful substances were being in- 
troduced rapidly into the workplace.^^ 

Following a series of compromises both within and between the two 
houses of Congress, the Occupational Safety and Health Act 
(OSHAct) was passed and signed into law in December 1970. In the 
statement of findings and purpose for the act, the Congress found : 

That personal injuries and illnesses arising out of work situations impose a 
substantial burden upon, and are a hindrance to, interstate commerce in terms 
of lost production, wage loss, medical expenses, and disability compensation 

The goal of the act was to "assure so far as possible CA^ery working man 
and woman in the Nation safe and healthful working condi- 
tions * * *." "^^ The report of the Senate Labor and Public Welfare 
Committee stated more succinctly that the purpose was to "reduce the 
number and severity of work-related injuries and illnesses." ^^ 

Provisions of the Act 

Standard Setting — The primary method by which the act seeks to 
insure safe and healtliful working conditions is through the promul- 
gation and enforcement of standards and regidations by the Secre- 
tary of Labor. Three classes of standards are authorized. The first, in- 
terim standards, were intended to a]>ply during the first 2 years after 
the Act took effect. Lender this procedure, the Secretary of Labor was to 
issue "as soon as practicable'' standards drawn from existing Federal 
regulations or from "national consensus standards,*' such as those es- 
tablished by the American National Standards Institute. In the case of 
conflicts, the "Secretary shall promulgate the standard which assures 
the greatest protection * * *." "^ 

The provisions for issuing permanent standards under the act re- 
quire that the Secretary of Labor follow more formal procedures 
under the Administrative Procedure Act. Briefly, the Secretary may 
appoint an advisory committee and must allow for comments, objec- 
tions, and a public hearing after publication of the proposed rule. 
Time limits for the various stages are specified in the act. In setting 
standards dealing with "toxic materials or harmful physical agents," 
the Secretary is directed to "set the standard which most adequately 
assures, to the extent feasible, on the basis of the best available 
evidence, that no employee will suffer material impairment of health 
or functional capacity even if such employee has regular exposure 

«• Smith, The OSIIAct. pp. 6-7. 

«s BNA. Job Safety and Health Act of 1070, pp. 13-14. 
«p OSHAct. sec. 2(a). 
"^Ihid., sec. 2(b). 

"^U.S. Senate Committee on Labor and Public Welfare. Lecislatlve History of the 
Occupational Safety and Health Act of 1970 (GPO, Washington, D.C.. 1971) | p. 141. 
"OSHAct, sec. 6ia). 


to the hazard dealt with by such standard for the period of his work- 
ing life." ^^ 

In those cases where employees are exposed to "grave danger," the 
Secretary is authorized to issue temporary emergency standards to 
take effect immediately upon publication. However, the Secretary is 
required immediately to initiate proceedings to establish permanent 

In addition to the employer's duty to comply with the specific 
standards established by the Secretary of Labor, the Act also imposes 
a general duty on each employer to "furnish each of his employees em- 
ployment and a place of employment which are free from recognized 
hazards that are causing or are likely to cause death or serious physical 
harm to his employees." ^^ Employees are also ordered to comply with 
regulations under the Act, but no provisions are made for enforcement 
or penalties. 

Enforcement — Under the OSHAct, Labor Department inspectors 
are authorized to conduct unannounced inspections of workplaces dur- 
ing regular working hours. The inspector is required to permit author- 
ized representatives of both labor and management to accompany him 
during the inspection. Where no authorized employee representative 
exists, the inspector must consult with "a reasonable number of 

If the inspection reveals a violation, the inspector must issue a writ- 
ten citation. In the case of "nonserious violations," the Secretary may 
propose penalties of up to $1,000 per violation. For "serious viola- 
tions," where "there is a substantial probability that death or serious 
physical harm could result," the Secretary shall assess a penalty, again 
not to exceed $1,000. The law also provides for civil penalties of up to 
$1,000 for failure to post required notices and for a fine of up to $10,000 
for willful or repeated violation. Failure to correct cited conditions 
within the abatement period specified by the Secretary can result in 
fines of up to $1,000 per day. In addition to civil penalties, the law 
provides for criminal penalties, both fines and imprisonment, in cer- 
tain cases, including willful violations resulting in death, unauthorized 
advance notice of inspections, and assaults on inspectors. 

Occupational Safety and Health Review Commission — Employers 
who wish to contest either the citation or the penalty may appeal to 
the Occupational Safety and Health Review Commission, an inde- 
pendent body established by the Act."^^ The Commission, consisting of 
three commissioners appointed by the President for staggered 6-year 
terms, has the power to vacate or modify the Secretary's citation or 
penalty followinjr a hearing. Decisions of the Commission m.ay be 
appealed in the U.S. Court of Appeals. If an appeal is not made to 
the Commission, the Secretary's citation and proposed penalty auto- 
matically become final orders of the Commission. 

The T^Tational Institute for Occupational Safety and Health — In 
order to provide the information needed by the Secretary of Labor to 
set stanrlards, the Act established the N'ational Institute of Occupa- 
tional Safety and Health (NIOSH) within the Department of 
Health, Education, and Welfare.'^^ In addition to providing criteria 

73 Ibid. 



'* Ibid., 



75 Tbid., 



76 Ibid. 




documents for standard setting, NIOSH is charged with conducting 
research related to occupational safety and health, compiling and 
periodically updating a list of toxic substances, carrying out surveys 
of hazards to workers, and providing training for personnel to carry 
out the act and for employers and employees. 

Role of the States 

Although the OSHAct provides for an unprecedented level of Fed- 
eral involvement in the promotion and regulation of occupational 
health and safety, it by no means preempts State involvement in the 
area. Committee reports in both the House and the Senate make clear 
that the act in no way affects State workmen's compensation laws, nor 
does it alter employee and employer rights under tort law.'^^ In addi- 
tion. States are permitted to enforce standards in those areas not cov- 
ered by Federal standards. Finally, States may regain responsibility 
for the "development and enforcement therein of occupational safety 
and health standards" covered by Federal reguhition by submitting 
an acceptable plan to the Secretary of Labor. '^^ The key requirements 
for State plans include the establishment of State standards "at least 
as effective" as Federal standards covering the same issue, and assur- 
ances that adequate funds and personnel will be provided for enforce- 
ment. The Secretary of Labor is authorized to reimburse States for 
up to 90 percent of the costs of developing such plans, and up to 50 
percent of the costs of administration and enforcement.'^* 

Record Keeping — Prior to passage of the OSHAct, data on occupa- 
tional injuries and illnesses were collected by the Bureau of Labor 
Statistics under the so-called ZlG.l standard. Participation of firms 
was incomplete and voluntary, and it was widely agreed that small 
firms in particular were under-represented. In order to improve the 
information available, the OSHAct requires employers to maintain 
records on "work-related deaths, injuries and illnesses other than minor 
injuries requiring only first aid treatment * * *." The Act also directs 
the Secretary of Labor to "develop and maintain an effective program 
of collection, compilation, and analyses of occupational safety and 
health statistics." ^° 

Othei' Provisions — The Act also provided for the establishment of 
two additional groups. The Xational Advisory Committee on Occupa- 
tional Safety and Health (XACOSH) , to be "composed of representa- 
tives of manngement, labor, occupational safety and occupational 
health professionals, and of the public," was to "advise, consult with 
and make recommendations to the secretaries of Labor and HEW on 
the administration of the Act." ^^ The Act also established the National 
Commission on State Workmen's Compensation Laws to examine 
existing compensation laws to "determine if such laws provide an ade- 
auate, prom])t, and eouitnble system of compensation for injury or 
death arising out of or in the course of employment." ^^ Members of the 
Commission were to be appointed by the President, and were to sub- 
mit a report to the Congress by July 31, 1972. 

■'"T'.S. Sonnte. Lpcislative History, p. 162. 

"'^OSITArt. spo. 18. 

™ Tl>if'., spc. 23. 

("i Tbif1.. sec. 24. 

" Tbifl., spc. 7. 

»- Ibid., sec. 27. 

8.3-944—78 14 

OSHA's Performance 


Five new Federal Agencies were formed to implement the OSHAct : 
The Occupational Safety and Health Administration (OSHA) with- 
in the Department of Labor; the National Institute of Occupational 
Safety and Health (NIOSH) within the Department of Health, Edu- 
cation, and Welfare; the Occupational Safety and Health Review 
Commission (OSHRC) ; the National Advisory Committee on Occu- 
pational Safety and Health (NACOSH) ; and the National Com- 
mission on State Workmen's Compensation Laws (NCSWCL). The 
NCSWCL, in accordance with its charge under the OSHAct, ceased 
to exist 90 days after it issued its final report in July 1972.®^ The 
other four agencies continue in existence, with OSHA bearing the 
primary responsibility for carrying out the provisions of the Act. 

Overview of OSHA 

OSHA is headed by the Assistant Secretary of Labor for Occupa- 
tion Safety and Health, a position created by the OSHAct. During 
its first 7 years of existence, the agency has gone through several 
reorganizations and changes in leadership. Figure 1 shows OSHA's 
organization chart as of January, 1978. OSHA's current head, Dr. 
Eula Bingham, came to her post from the University of Cincinnati, 
where she taught in the medical school and was associate director of 
the Institute of Environmental Health. Her immediate predecessor, 
Dr. Morton Corn, who was also an academic specializing in occupa- 
tional and environmental health, held the post from December, 1975 
to January, 1977.«* 

83 NCSWCL Report. 

^' On Bingham's background, see "Life-threatening Hazards Targeted by Agency Leader," 
Times Standard, Eureka, California, May. 15, 1977. On Corn's background, see J. W. Singer, 
"Labor Report/New OSHA Head May Signal Change in Agency's Approach," National 
Journal (December 27, 1975). 
































•s g-.1 



TJ O > 

.iJ N 

O fc. CU < 

O w 1 

t-i .13 





a; -H 
Q g 


ff. of 



G < 

C (0 

u o 

O V4 

O CI. 

— PI 
^ 8 

o c. 

0) w 
m .-1 
o u 
o c 
M u 



Off. of Environmental, 
Inflationary, S. 
Economic Impact 






^ s 

o c 

t-« 4J 

O '</] 

o 1 




l-t ^ 



10 U) 

>, 4J 


^ D 


Cm < 


M W 













o < 

M O 

o c 

w o 


(In thousands of dollars] 

Fiscal year- 

1971 1972 

budgeted budgeted 1973 1974 1975 1976 1977 

and actual and actual actual budgeted actual actual budgeted 

Standards 1,062 

Training 455 

State programs 5,764 

Federal enforcement 6,063 

Statistics 758 

Executive direction and administration... 1, 082 












18, 897 


27, 080 

25, 080 

41, 000 

32, 969 



24, 161 


42, 177 

59, 752 



4, 814 











Total budget 15,189 35,684 69,274 70,408 102,327 114,980 130,333 

Sources: 1971-75, Ashford, Crisis, p. 240. 1976-77, memorandum from M. J. Boile, Congressional Research Service, 
Library of Congress, Jan. 6, 1977, 


As described earlier, the OSHAct specifies three different proce- 
dures under which standards can be adopted. On May 29, 1971, only 
5 months after the OSHAct was signed into law and a month after 
it took effect, OSHA issued interim standards drawn from existing 
Federal and national consensus safety and health standards. Roughly 
4,400 standards were adopted, almost half of which had been estab- 
lished by the American National Standards Institute or the National 
Fire Protection Association. Although the vast majority related to 
worker safety rather than health, threshold limit values (TLV's) 
were included for approximately 400 substances in common indus- 
trial use.^^ 

During OSHA's first 4 years of existence, only four major sets of 
standards were promulgated: three health standards (vinyl chloride, 
asbestos, and 14 carcinogens) and a group of safety standards cover- 
ing mechanical power presses.^^ Under Corn, OSHA began to 
accelerate the development of standards, with a particular 
emphasis on health standards. In May 1975, the head of OSHA's 
Office of Standards Development estimated that roughly 98 percent 
of OSHA's standards, as measured by page lengt.h, were safety rather 
than health related.^^ OSHA and NIOSH are working on a "crash" 
program to convert all 400 of the TLV's to complete standards.^* 
Other major new standards packages include requirements for roll- 
over protection and guarding for aoricultural equipment and stand- 
ards to limit coke oven emissions. OSHA is also considering a reduc- 
tion in permissible noise level s.^^ 

One of the most frequently voiced criticisms of OSHA standards 
is that they are too numerous and complex for employers to under- 
stand. An official of the National Association of INIanufacturers testi- 
fied at a 1974 House hearing that: 

Largre employers are still confused and medium and small employers continue 
to he overv'helmed by the volume and complexity of OSHA standards and 

^- Smith. T!ie OSHAct, pp. 9-10. 

^' Tbid.. p. 10. 

8'^ "No rf^oisions," Job Safety and Health. 3 {"S^ixv lOTo). 22-2'^. 

^'^G. rifick. "The First of 400." .Tob Safety and Health. 3 (.Tune 1975), 4-10. 

^9 TTearlnjrs were held in the fall of 197fi on tlie proposed standard, inr-liiding considera- 
tion of n t!"ht(>nin£r from the cnrront 90 dBA 8-Jionr Aveifrhted standard to one of So dBA. 

s" Sfatemont of M. P. Stinton. U.R. Honse. Committee on Edncation and Labor. Select 
Subcomm.ittee on Labor, Occupational Safety and Health Act of 1970 (Oversight and 
Proposed Amendments), Hearings (^Sd Cong., 2d sess. ), p. 321. 


In an effort to make its standards more easily accessible to employ- 
ers, OSHA has issued a number of pamphlets explaining standards 
in simpler terms, including alphabetical digests reputed to cover most 
of the standards applicable to general industry and to the construc- 
tion industry.^^ 

Many of the criticisms of OSHA standards have been directed at 
the consensus standards adopted en masse in 1971. Some of those 
standards, such as a requirement that toilet seats be split and not 
round,^^ related only peripherally to worker safety or health. Others 
had long been obsolete, one of the more extreme examples being a 
prohibition against ice in drinking water, which dated from the time 
w^hen ice was cut from frozen polluted rivers and lakes.^^ Some of 
the standards adopted in 1971 have since become obsolete as OSHA 
has failed to keep pace with updates made by the consensus 

Since 1971, OSHA has succeeded in eliminating or revising some of 
the more troublesome consensus standards. Currently a program is 
underway to review and revise approximately 15 percent of the 
remaining consensus standards.^^ In addition, a task force appointed by 
President Ford completed the revision of an additional 10 percent of 
the consensus standards.^^ 

Few would disagree with the proposition that OSHA should strive 
to revise or eliminate obsolete or irrelevant standards, and attempt 
to write standards in a simpler, easier to understand manner. Con- 
siderably less agreement can be secured, however, on the level of pro- 
tection which OSHA should strive for in setting standards, and the 
extent to which economic considerations should influence the strin- 
gency of the standards chosen. 

The language of tlie OSHAct provides no unambiguous criteria for 
determining the degree of protection which standards should afford. 
In few cases, unfortunately, is there a particular level which can be 
labeled unambiguously as "safe and healthful." In most cases, as the 
standard is made more restrictive, the level of danger continues to 
derrease; at no point, short of banning the substance or prohibiting 
altogether the potentially hazardous activity, does the risk disappear. 
Thus a decision must be made as to what level of protection shall be 
achieved, or at least strived for. 

The OSHAct makes no ex])licit provision for the consideration of 
economic costs, although the level of protection to be afforded workers 
is qualified by such phrases as "so far as possible" and "to the extent 
feasible." "Feasibility" has been interproted as reforrinir primarily 
to technical feasibility: that is, whether the standard can be met with 
existing technolocrv.^^ The role of economic feasibilitv in determinin<r 

SI "Standards Disrpst : Conoral Industry" COcinA 2201) mid "Standards Dicrost : Con- 
stDiotion" (OSHA 2202) are advertised' and described in various issues of Job Safety 
and Health. 

P2in72 CFR 1010.141(0) (3) (ii). 

"''1072 CVn 1010.141(b) (lUii). The reason for this re-ulatlon is civen in Ashford, 
Trisis. p. 24,9. The 107."^ edition of the CFR merely requires that ice in drinking water 
be nifide from potable water. 

^* See "New Procedures for Safety Stnndards Revision." Job Safetv and Henlth, 4 (June 
1070). p. 0. v^hich nntes that OSTT.V has not kept pace with revisions of the National 
Fire Protection Association standards. 

"'The standards included in tlr's revision .affect walkin£r-workincr surfaces, fire orotec- 
tion. and anhydrous ammonia. I^S. Dopartment of Labor pross release, Apr. 10, 107(>. 

""J. W. Sincror. "New OSTTA Task Force — Political Payoff or False Alarm?" National 
Journal (July 10. 107(>), pp. 07.'^-07r). 

^ Smith. Tlie OSHAct. p. la. notes that virtually no discussion of the meaning o£"feasi- 
biilty" occurred in Congress when the OSHAct was passed. 


the validity of standards appears to be limited to extreme cases, where 
imposition of the standard would drive a substantial proportion of the 
firms in an industry out of business. In one major case, the court of 
appeals held that : 

Practical considerations can temper protective requirements. Congress does 
not appear to have intended to protect employees by putting their employers 
out of business — either by requiring protective devices unavailable under exist- 
ing technology or by making financial viability generally impossible." 

Although this decision does permit the inclusion of economic con- 
siderations in determining the validity of a standard, it most defi- 
nitely does not call for any attempt to balance the costs of complying 
with a standard against the benefits derived. In a number of cases, 
the OSHKC has ruled that the high cost of meeting an OSHA 
standard relative to its improvement in the level of safety is not a 
valid defense against a citation.^^ 

Despite the narrow interpretations of economic feasibility discussed 
above, it is clear that OSHA has given at least some consideration to 
costs^ in setting standards. OSHA is required, under the terms of a 
Presidential order applying to executive branch agencies generally, 
to prepare inflationary impact studies for major new standards.^°° 
Those studies, despite their titles, deal not so much with inflation per 
se as with the general economic impact of standards. Unfortunately, 
while costs are tabulated in considerable detail, little effort is made to 
derive even a crude measure of the benefits.^^^ Furthermore, typically 
no effort is made to compare the costs and benefits, even in as simple 
a fashion as expressing the costs in a per-worker-protected basis. 
For example, the inflationary impact statement (IIS) for the 
coke-oven emissions standard estimates that total annualized costs 
will be in the range $241 million to $1,280 million. About 29,600 
workers are employed in coke-oven departments, and the IIS esti- 
mates that the excess mortality due to such employment is in the range 
26 to 240 deaths per year."- If one divides the cost estimates by the 
number of workers employed, the estimated cost per-worker- 
protected-per-year is between $8,100 and $43,200. These amounts 
should be compared with the average wage of workers in blast 
furnaces and steel mills, which is about $13,000 per year."^ If one 
divides costs by lives saved, assuming that compliance with the stand- 
ard will eliminate all excess mortality, the range is between $1 million 
and $49.2 million per life saved. A critical review of the coke ovens 
IIS bv the Council on Wage and Price Stability concludes that a more 

88 Industrial Union Departments. AFL-CIO v. Hodgson, 1 OSHC 1639, Apr. 15, 1974, 
quoted in Smith. The OSHAct. p. 16. 

»» Smith, The OSHAct. p. 17. 

100 Executive Order 11821, "Inflation Impact Statement," Nov. 27, 1974. 

'01 In many cases, no effort whatsoever is made to assess the henefits. See a series of 
Inflationary impact assessments issued by OSHA on May 25, 1976, covering proposed 
standards for 2-butanone, cyclohexanone, ethyl butvl ketone, hexone, methyl (n-amyl) 
ketone, and 2-pentanone. Each of these studies is designed to show that the economic 
thresholds for preparing a full IIS are not exceeded. The difficulty, of course, is that 
while the costs may be small (compliance costs less than $100 million in any year, enerery 
demand less than the equivalent of 25.000 bbls per day. etc.). the benefits for OSH may 
also be very small. We stress again the importance of looking at costs in relation to 
benefits, not as absolute amounts. 

102 U.S. Department of Labor, Inflationarv Impact Statement: Coke Oven Emissions, 
Occupational Safety and Health Administration (Feb. 27, 1976). One of the reviewers of 
an earlier version of this report notes that OSHA and the contractor preparing the coke 
oven IIS resisted including any data on benefits, and did so only lat the Insistence of 
another office wtihln the Department of Labor, 

103 Ibid., p. SO. shows average earnings of !R25R.75 per week for workers in blast fur- 
naces and steel mills, which translates to about $13,000 per year. 


realistic estimate is in the range $4.5 million to $158 million per life 
saved. The Council contrasts these estimates with the $240,000 per 
life used by the National Highway Traffic Safety Administration in 
estimating the societal costs of motor vehicle accidents and the $1 
million used as a high estimate by Consumers Union in evaluating 
the proposed lawn mower safety standard for the Consumer Products 
Safety Commission. ^^"^ 

Two more specific issues arise in connection with the question of 
trading off costs and benefits in setting standards: (1) The role of 
personal protective equipment in reducing exposure to hazardous 
conditions; and (2) the extent to which OSHA should modifs^ stand- 
ards on an industry-by-industry or firm-by-firm basis. OSHA's gen- 
eral policy has been to demand that employers meet standards through 
engineering controls, such as ventilation systems to remove airborne 
contaminants, or more rarely, through administrative procedures, 
such as the rotation of workers in hazardous areas, rather than 
through the use of personal protective equipment for employees, such 
as earplugs or respirators.^°^ Such devices typically have been allowed 
only when existing technology' does not permit an engineering solu- 
tion, or as a temporaiy protective measure until more permanent 
solutions can be implemented. OSHA's stand on this issue has been 
criticized severely by management, particularly with regard to the 
noise standard, where relatively inexpensive earplugs or muffs can 
provide workers with the same protection as far more costly chansfes 
in the design of plants and equipmont.^^^ In a ca-e before the OSHRC, 
the Continental Can Co. presented evidence ^^howing that to achieve 
the 90 dBA noise standard would cost $32 million in capital expendi- 
tures for engineering controls, as opposed to only $100,000 for per- 
sonal protective devices. In that case, the Commission held that OSHA 
had not demonstrated the economic feasibility of engineering controls 
and permitted Continental Can to use personal protective devices. 
The Commission made it clear, however, that engineering controls 
are to be used to the extent that thev are feasible, both economically 
and' The Council on AVaire and Price Stability ha^ also 
sugirested that the provision of respirators for coke-oven workers 
would provide a relatively inexpensive alternative to the engineering 
and administrative controls mandated by the OSHA standards.^°^ 

OSHA's primary argument against personal protective equipment 
has been that workers often fail to use such devices because they are 

104 T. F. Morrall III. "Statement on Behalf of the Council on Wajre and Price Stability, 
tpstiniony heforo t!ie Ocoupafional Safety and Health Administration, Exposure to Coke 
Ovon Emission : Proposed Standard." docket No H-017 \ 

1^5 Smith. The OSHAct, pp. 4.3-49. 

io«Bolt. Beranek. and Newman. Inc.. "Economic Impact Analvsis of Pronospd Noise 
control Rpjrnlation." a report prepared for the U.S. Department 'of Lahor. Occupational 
Nafptv and Health Administration. April 12. 1976, estimates that an So-dBA standard 
could be achieved through the use of hearing protection at an annual cost of $4.3 million. 
as opposed to a cost of $18.5 billion in capital costs alone if engineering controls are 
required. o & 

i<" Cited In N. A. Ashford et al.. "Economic/Social Impact of Occupational Noise 
iixposure Regulations." Testimony presented at the OSHA hearings on the eronomic 
impa^ct of occupational noise exposure, Washington, D.C., September 30, 1976, pp. 7-4 

^n-i^^ ^^^ ^^^^ ^''^^' ^^^ OSHRC ruled that the Turner Company (No. 363n. Aucust 24, 
19. b) must Install engineering controls to reduce noise exposure. In his dissent. Com- 
missioner Moran noted that while the total cost of engineering controls was only ?25.000. 
the post per worker was about the same as in the Continental Can case, $3,000. Occupa- 
tional Safety and Health Reporter. August 26. 1976. p 35^0 

los Morrall, "Statement on Behalf of the Council on Wage and Price Stability," p. 12. 


uncomfortable to wear or interfere with the employee's activities. 
Organized labor has been particularly vociferous in its opposition 
to the use of personal protective equipment. A principal source of 
their concern is that firms may have difficulty enforcing a require- 
ment that workers use the equipment provided. However, given that 
the personal protection approach may offer tremendous cost savings, 
OSHA! should give it active consideration for some problem areas, 
noise probably i3eing the most significant example. 

Most OSHA standards apply across all industries and firms, regard- 
less of variations in costs and benefits. Thus, for example, metal 
stamping plants and lumber mills are subject to the same noise stand- 
ard that applies to retail stores and financial offices. As discussed 
earlier, efficiency, in the sense of maximum protection per dollar spent, 
is achieved when the ratio of marginal costs to marginal benefits is the 
same at each location. A recent study by Nicholas Ashford et al. sug- 
gests that uniform noise standards will lead to widely varying cost/ 
benefit ratios in different industries. For example, in the case of an 
85-dBA standard imposed over 5 years, the ratio of the present value 
of the costs to the number of person-years of hearing impairment 
prevented over 45 vears ranges from $300 in apparel and other textile 
products (SIC 23) to $2,680 for tobacco manufacturers (SIC 21), 
with a mean of $840.^°^ This wide variation suggests that it would 
be possible to reduce substantially both costs and the number of 
workers with hearing impairment by adopting a somewhat relaxed 
noise standard for high cost/benefit industries, such as tobacco manu- 
facturers, and a tighter standard in those industries with a lower 
cost /benefit ratio, such as apparel and other textile products. Al- 
though it would be impracticable for OSHA to set separate standards 
for each industry, it might well be desirable for it to consider relaxing 
particular standards for industries with much greater than average 
cost/benefit ratios. 

Another way in which OSHA could reduce the costs of compliance 
would be to increase the use of "grandfather clauses" in its standards; 
that is, make standards applicable only to those plants or pieces of 
equipment manufactured after the standard becomes effective. Un- 
fortunately, it is often far more expensive to modify existing facilities 
and equipment than it is to incorporate those requirements into the 
original design. For example, OSHA standards specify that certain 
railings be constructed of 1.5-inch pipe. For firms installing new 
railings, the cost of meeting this requirement is minimal. However, 
for other plants which currently have in place railings constructed 
of smaller diameter pipe, the cost is far greater, as the entire railing 
must be replaced.^^^ 

In a few cases. OSHA has inserted "grandfather clauses" in its 
standards. The rollover protection requirement for farm tractors pro- 
vides a recent example where existing eouipment has been exempted.^^^ 
In that case, however, it appears that the exemption was based not so 
much on extra costs as on the fact that many older tractors do not 

i<» Ashford et al., "Economic/Social Impact of Occupational Noise Exposure Regula- 
tions." table 6.1. 

^" See testimony of C M. Skinner. U.S. Congress. House. Permanent ^('^^ct Committee 
on Small Business, Subcommittee on Environmental Pi-oblems Aflrectin<r Small Business, 
The Effects of the ArJmiPistratlon of the Occupational Safety and Health Act on Small 
Business, Hearin.cs (93d Cong.. 2d Sess.), p. 173. 

"i See "ROPS Required on Tractors," Job Safety and Health, 3 (July 1975), p. 2. 


have frames lending themselves to the attachment of rollover cages. 
Although such "grandfather clauses'' do reduce temporarily the pro- 
tection afforded workers.^ the high costs of modifying existing plants 
and equipment make them desirable in many cases. 


Measured either in terms of dollars or manpower, enforcement is 
the major activity of OSHA. In fiscal year 1977, OSHA budgeted 
$57.6 million for "Federal enforcement, about 44 percent of OSHA's 
total budget of $180 million and roughly 61 percent of its total ex- 
penditures excluding grants to States. Of the 2,717 positions author- 
ized for OSHA in 1977, 2,216, or 82 percent, were allocated to Federal 

The total number of inspections conducted by OSHA compliance 
officers has grown rapidly, from 14,500 during 1971, when the act 
was in effect for only part of the year, to 88,801 in calendar year 1975. 
Data through April, 1976 show inspections in that year running about 
17 percent ahead of 1975's pace."^ Even at the current rate, however, 
only about 2 percent of the approximately 5 million workplaces cov- 
ered by the act will be visited by a Federal inspector in any one year. 
Inspections are also conducted, of course, by State inspectors in those 
jurisdictions operating under approved plans. Ashford estimates that 
in 1975, between 1,500 and 1,800 State inspectors were active under 
such plans."* 

Given its inability to inspect more than a small fraction of the 
workplaces under its jurisdiction in any 1 year, OSHA was forced 
to establish inspection priorities from the very beginning. OSHA 
initially selected five high-injury-rate industries — longshoring, lum- 
ber and wood products, roofing and sheet metal, meat and meat prod- 
ucts, and miscellaneous transportation manufacturing (primarilv mo- 
bile homes) — for the Target Industry Program (TIP), and five 
widespread health hazards — asbestos, cotton dust, silicon dust, lead, 
and carbon monoxide — for the Target Health Hazards Program 
(THHP)."^ By the end of 1972, virtually all TIP firms with more 
than 20 employees had been inspected."^ Inspections under the THHP 
were far more limited, comprising only 2.8 percent of all inspections 
conducted from July 1974 through ]\[ay 1975."' Other than noise, few 
health standard violations are cited during inspections not carried 
out under the THHP. In large part this reflects the composition of 
the OSHA inspection force, which is heavily weighted toward safety 
specialists as opposed to industrial hygienists. The mix is now changing 
slowly as more industrial hygienists are hired and trained under 
policies initiated by Corn."^ 

The TIP and, to a lesser extent, the THHP have been replaced es- 
sentially by the National Emphasis Program (NEP). The first in- 
dustry selected for the XEP is foundries, which expose their workers 
to a variety of both safety and health hazards."^ As part of NEP, 

"2 Djita supplied by Leslie Beckenbach. OSHA, Nov. 9, 1976. 

"3 Job Safety and Health. 4 (August 1976), p. 3. 

"* Ashford. Crisis, p. 226. 

"^The President's Report on Occupational Safety and Health, 1972 annual report. 

ii« Smith. The OSHAct. p. 68. 

"^ Ashford. Crisis, p. 257. 

""Interview with Morton Corn, Assistant Secretary for Occupational Safety and 
Health. U.S. Department of Labor. An?. 20. 1976. 

"9 The April 1976 issue of Job Safety and Health contains several articles on the NEP 
for foundries. 


OSHA has worked closely with the relevant trade organizations, and 
its inspectors have been receiving special training in the particular 
hazards likely to be found in foundries. It is expected that foundries 
inspected under the NEP will be visited by a team of inspectors that 
includes at least one safety specialist and an industrial hygienist. 
These teams will spend considerably more time at each workplace 
than is usual, and will attempt to evaluate a firm's overall safety and 
health program, in addition to checking for violations of specific 
standards.^^^ Current plans call for gradually developing additional 
NEP's for other industries. In some cases, regional administrators or 
area directors have selected certain industries or major projects, such 
as construction of the Metro system in Washington, D.C., for local 

In addition to NEP, OSHA has a four-tiered hierarchy of inspec- 
tion priorities : ^^^ 

1. Employee complaints of imminent danger situations. 

2. Investigations of fatalities and catastrophes. 

3. Other employee complaints. 

4. General schedule inspections. 

As shown in table 2, OSHA's distribution of inspections has re- 
mained fairly steady since 1974, although investigations of employee 
complaints increased somewhat during January-April 1976. Com- 
plaint investigations are expected to become even more important 
since OSHA has altered its procedures and acts on virtually all em- 
ployee complaints, not just on those filed formally on the appropriate 
OSHA form.^23 

Ninety percent of general schedule inspections are now supposed to 
be conducted on the so-called worst-first basis. Within each State 
industries are ranked by the average number of injuries per establish- 
ment, which is simply the industry's injury rate times the average 
number of employees per establishment. Thus an industry with a rela- 
tively low injury rate, but which is composed of large establishments, 
may be ranked ahead of an industry with a higher injury rate, but 
which is made up of smaller establishments. Within each industry, 
establishments are ranked by their expected number of injuries, which 
means, since industrywide average injury rates are used, that they are 
ranked by number of employees.^^^ 

The worst-first approach assumes that the more dangerous the 
workplace (in terms of expected number of accidents) the more 
productive (in terms of accidents avoided) an OSHA inspection is 
likely to prove. Obviously, factors other than an establishment's ex- 
pected number of accidents might be employed to compute a produc- 
tivity measure for OSHA inspections. For example, it would be help- 
ful to have an index that indicates how much accidents are reduced 
when particular violations are abated. However, to identify the ap- 

120 w. Carroll, "The New Team Approach to Foundry Inspection," Job Safety and 
Health, 4 (August 1976), 12-17. 

1-1 Interview with Gilbert Saulter, Regional Administrator for OSHA, Boston, Mass., 
and former Area Director, Washington, D.C., August 24, 1976. 

-^ Interview with Harlan Holdefer and Joe Nolan, Field Performance Evaluation 
Team, OSHA, Washington, D.C., August 20, 1976. 

1^ Interview with Saulter. See also, "Employee Complaints to Receive New Treatment," 
Job Safety and Health, 4 (June 1976), p. 2. As this report was going to press OSHA 
sources reported that in the third quarter of 1976 complaints accounted for 30.2 percent 
of total inspections, well above their earlier share. 

1"* The authors' understanding of the "worst-first" strategy Is based on conversations 
with several OSHA officials, particularly Gilbert Saulter, and examination of materials 
sent to the New England regional office of OSHA. 


propriate factors would require more information than is presently 
available. The general concept of worst-first, then, may well be a 
reasonable one for targeting inspections. 

(In percent] 

Type of inspection 





















Source: 1974 and 1975: Job Safety and Health, May 1976, p. 3. 1976: Job Safety and Health. August 1976, p. 3. 

Several objections may be made to the specific worst-first strategy 
currently employed. That strategy assumes that an inspector's ex- 
pected productivity depends only on the expected number of accidentvS 
in the establishment inspected. Thus, it is assumed, an inspectx)r will 
be just as effective if he inspects a relatively safe large workplace with 
1,000 employees and an accident rate of 2 injuries per 100 man-years 
as he will if he inspects a much smaller, relatively more dangerous 
\yorkplace with 100 employees and an accident rate of 20; l>oth estab- 
li.shments on average will experience 20 injuries per year. This con- 
clusion, however, requires two dubious assumptions. First, it must be 
assumed that it takes no longer to inspect the larger establishment 
than the smaller one. Although there are many fixed time inputs re- 
quired regardless of the size of the establishment, such as travel and 
report preparation, some time requirements do increase with the size 
of the firm. The second assumption made implicitly is that inspections 
will lead to the same proportionate reductions in accident rates in low- 
in jury^-rate establishments as in high-rate establishments. Intuitively 
it would seem more likely that establishments with low injury rates 
already will have taken many of the more obvious steps to reduce acci- 
dents, and that further reductions will be far more difficult to achieve. 

A firm need not actually be inspected, of course, in order to come 
into compliance with OSHA standards. Some employers will comply 
simply because it is the law. Where standards apply to equipment, 
firms may come into compliance when they replace obsolete equipment ; 
although OSHA standards apply to the firms using equipment and not 
to those producing it, manufacturers are unlikely to offer new equip- 
ment which will be in violation. 

The fines levied for violations of OSHA standards also provide a 
potential incentive for voluntary compliance prior to an inspection. Op- 
position to proposals to eliminate penalties for fii'st-instance violators 
has been based on the argument that if employers knew that they could 
not be fined unless they failed to abate the violations cited during an 
inspection, few^ would bother to comply voluntarilv. In fact, it appears 
that few finns do comply prior to inspections. During 1974, only 23 
percent of the firms inspected were "in compliance." During 1975 and 
the first 4 months of 1976, the figure was still lower, 21 percent.^-^ 


Examination of the dollar penalties proposed by OSHA makes clear 
that the costs of noncompliance are extremely low. As shown in table 3, 
the average proposed penalty per violation has ranged from $20.60 in 
1971 to $31.76 in 1976. Most firms are cited for more than one violation^ 
but the average per citation has still been extremely low, reaching a 
peak of _ $163 in the first 4 months of 1976. The fines levied by States 
operating approved plans have averaged even less.^^^ These small fines, 
coupled with the low probabilities of inspection noted earlier, give 
firms little incentive to comply prior to an inspection. Once a firm has 
been inspected and cited, however, the incentive to comply rises dra- 
matically; the probability of reinspection is relatively high and the 
penalties for failing to abate violations are substantial, up to $1,000 
per day.^27 








Average dollar per 
violation.. _ 

Average dollar per 

20. 60 
77. 63 

130. 59 

140. 88 

126. 32 

144. 23 


Source: 1971-73: President's Report on Occupational Safety and Health, annual report for 1973, 'p. 5. 1974-75: Job 
Safety and Health, 4 (May 1976), p. 3. 1976: Job Safety and Health, 4 (August 1976), p. 3. 

The small penalties imposed on cited firms by OSHA partly reflect 
the fact that the overwhelming majority of violations are classified as 
"nonserious." During calendar year 1974, fully 98.5 percent of the vio- 
lations were "nonserious," only 1.2 percent were rated "serious," and a 
minuscule 0.3 percent fell in the category "imminent danger, willful 
or repeated." ^2^ Furthermore, classification criteria have been far 
from uniform in the different regional offices; a study of citations 
issued during fiscal year 1974 found that the percentage of total viola- 
tions rated as serious ranged from 0.3 percent in region V (Illinois, 
Indiana, Michigan, Minnesota, Ohio, and Wisconsin) to over 3.5 per- 
cent in region X (Alaska, Idaho, Oregon, and Washington) .^^^ Na- 
tionally, the proportion of violations classified as serious has been ris- 
ing; during the first 4 months of 1976, 2.2 percent of the violations 
found were labeled serious, 96.4 percent nonserious, and 1.4 percent as 
imminent danger, willful, or repeatcd.^''^ Dr. Corn directed a re- 
examination of the classification system, with the aim of increasing 
the proportion of serious violations, while reducing the minimum pen- 
alty which an inspector can propose for a serious violation.^^^ OSHA 
officials say that inspectors have been reluctant to classify violations 
as serious in part because of the relatively high penalties associated 
with such citations. 

126 According to data in .Tob Safety and Health, 3 (October 1975), p. 2, the average 
fine per violation in the 23 State plans in operation during January-March 1975, was 

i27 0SHAct, section 17(d). 

i2« Job Safety and Health, 4 (May 1976), p. 3. 

i^U.S. Conjrress, Senate Committee on Labor and Public Welfare, Occupational Safety 
and Health Act Review, 1974. Hearings (9.3d Cong. 2d sess.). p. 993. 

i''o Job Safety and Health, 4 (August 1976), p. 3. 

131 Interview with Corn. 


State flans 

The OSHAct gives States the right to recover control of occupa- 
tional safety and health regulation if the Secretary of Labor deter- 
mines that their plans meet certain requirements specified in the act, 
the most important of which is that they provide for standards which 
are "at least as effective" as the corresponding Federal standards.^^^ 
By the end of 1974, all but five States (Kansas, Louisiana, Nebraska, 
Ohio, and South Dakota) had submitted plans at one time or an- 
other.123 However, nine States which had submitted plans, including 
several which had secured approval and had begun operation, with- 
drew their plans by mid-1975.^2* In January 1977, 28 States (plus 
the Virgin Islands) had approved plans. In many of those States, 
regulation of safety and health in maritime industries has remained a 
Federal responsibility.^^^ OSHA continues to monitor State plans fol- 
lowing approval in an effort to assure that they are implemented ap- 
propriately. In some States without their own plans, OSHA has con- 
tracts that pay 50 percent of the salaries of State inspectors to 
provide supplementary enforcement of Federal standards.^^^ 

Organized labor has consistently opposed the return of OSH regu- 
lation to the States, arguing that the OSHAct was passed largely in 
response to th^ States' poor performance in the area. The Secretary 
of Labor has been attacked for failing to gain sufficient assurances 
that approved plans will indeed be "at least as effective" as OSHA. 
For example, George Meany, president of the AFL-CIO, has attacked 
State plans as "unsecured promissory notes of future perform- 
ance." ^^^ Labor opposition has been a key factor in the withdrawal of 
several State plans."® That opposition and, probably more impor- 
tantly, the tight budget conditions in many States make it unlikely 
that State participation will exjxand significantly. The 50 percent ceil- 
ing on Federal support for the operation of State plans, as compared 
to the 90 percent subsidy for their development, makes an increasing 
number of States reluctant to take over from OSHA. 

Related agencies 

OSHRC— Under the OSHAct, the Secretary of Labor (in practice 
OSHA) may only propose penalties. Penalties proposed by OSHA 
only become collectable if they are upheld by the OSHRC, or if the 
employer makes no move to contest the citation within 15 days of re- 
ceiving it, in which case the citation and the proposed penalty auto- 
matically becomes a final order of the Commission."^ 

The OSHRC consists of three commissioners appointed by the 
President subject to Senate approval. Cases appealed to the Commis- 
sion are heard first by one of the Commission's hearing examiners. 
These hearings are held as close as possible to the community in which 
the employer is located. The examiner's decision in a case is considered 

^3-'OSTTAot. section 18(c). 

1"-^ Tho Pr^'sident's Report, annual report for 1974, p. 40. 

i"^' Ashford. Crisis, p. 227. 

i:i!>rp|,p President's Keport. annual report for in7.S, pp. 24-25. 

136 The President's Report, annual report for 1974, p. 43. 

^•"" Ouofed in Ashford, Crisis, p. 229. 

13.S por f^ hvS^i discussion of labor's opposition to State plans see an Interview with 
Georjre Tnvlor, chairman of the AFT>-CIO StPndine Committee on Occupational Safety 
and Ileal fh and a member of NACOSH, "Critic and Adviser," Job Safety and Health, 
4 nr>v lOTH). pp. .•".:{- .'U. 

130 OSHAct, Section 10(c). 


final unless one of the three commissioners requests a review by the 
Commission itself ."° 

Employers appear to be contesting an increasing proportion of the 
citations issued by OSHA. In calendar year 1974, the number of cases 
contested was 2,348, or 4.0 percent of the citations issued in that year. 
The corresponding figures in 1975 were 4,056 and 6.1 percent. In the 
first third of 1976, the number of contested cases equaled 16.0 percent 
of the citations issued."^ 

NIOSH — NIOSH's primary task is to develop criteria documents 
on job safety and health hazards for OSHA to use as the bases for 
setting standards to protect workers. In preparing these documents, 
NIOSH relies primarily on reviews of existing studies. NIOSH's own 
research has been criticized for overemphasizing small-scale labora- 
tory animnl studies, at the expense of more relevant epidemiological 
studies.^*^ The Institute has also been criticized for the slow pace at 
which it issued criteria documents in its first few years (only 18 had 
been transmitted by July, 1974) ,"^ but the process has been greatly 
accelerated since then (by the end of 1976, 65 criteria documents had 
been sent to OSHA).^** The slow rate at which health standards have 
been promulgated appears to be due more to delays within OSHA than 
within NIOSH."^ When interviewed in 1977, OSHA's director of 
health standards development said that NIOSH criteria documents 
were being received faster than they could be turned into standards, 
although the quality of the documents is "a matter of continuing con- 
cern." Criteria documents cannot be turned into a standards ^'simply 
by tearing off the cover and stapling on a new one;" OSHA must take 
account of both engineering and, secondarily, economic feasibility 
in setting standards.^*^ OSHA reportedly improved its relations 
with NIOSH under Dr. Corn."^ 

NACOSH— The OSHAct specifies that NACOSH be composed of 
16 members — 12 appointed by the Secretary of Labor and 4 by the 
Secretary of Health, Education, and Welfare — chosen to represent 
labor, business, occupational safety and health professionals, and the 
public.^^^ The Committee meets approximately 6 times a year and has 
a very limited budget and staff.^*^ Its influence on OSHA and NIOSH 
appears to be minimal. One observer claims that "for the most part, 
neither OSHA nor NIOSH takes NACOSH very seriously." ^^^ The 
situation seems to be improving, however; under Corn, NACOSH 
established four subgroups on budget, compliance, policy and issues, 
and standards which, it is hoped, will allow the committee to "cover 
more adequately some of the elements of the program." ^^^ 

i«> OSHAct, section 12. 

i« Sources: 1974-1975: Job Safety and Health, 4 (May 1976), p. 3. 1976: Job Safety 
and Health, 4 (Ausrust 1976), p. 3. 

i*2Ashford, Crisis, pp. 300-301. 

it^TT.S. Senate, OSHAct Review, 1974, p. 1056. 

^*4 Memorandum from Bolle. 

^*s In a 1975 interview, then-director of OSHA's Office of Standards Development Daniel 
Boyd noted that "22 milestones" must be passed before a standard takes effect, "No Easy 
Decisions." pp. 23-24. 

i** Intervievtr with Grover Wrenn, Chief, Division of Health Standards Development,. 
OSHA. Washinerton. D.C., Aujjust 20. 1976. 

"' Interviews with Wrenn and other OSHA officials. 

1*8 OSHAct. pec. 7(a). 

"9Ashford, Crisis, p. 246. 

i«>Ibid., p. 302. 

151 "Critic and Adviser," op. cit, p. 33. 



Given the broad scope and ambitious goals of the OSHAct, one 
would expect to find substantial impacts in a variety of areas. In this 
section we attempt to quantify some of those impacts. Unfortunately, 
the data available are limited and piecemeal ; broad, firmly based con- 
clusions cannot be made. We divide our examination of OSHA's im- 
pact into three areas : (1) the extent to which the OSHAct has achieved 
its intended purpose, the reduction of injuries and illnesses among 
workers; (2) the financial burden imposed on firms meeting require- 
ments of the OSHAct: and (8) the impact of the Act on the ^'enoral 
economy, with special reference to its effects on inflation and 

Injuries and Illnesses 

The central purpose of the Congress in passing the OSHAct, as the 
language of the Act itself and its legislative history made abundantly 
clear, was to decrease the number of accidents and occupational ill- 
nesses suffered by workers. Thus any analysis of OSHA's effective- 
ness must begin with an attempt to determine what impact it has had 
on occupational injury and disease rates. Some ^lembers of Congress 
expected that impact to be very great, as suggested by the fact that 
during hearings held in 1972. one of the Act's authors. Congressman 
Steiger, made informal remarks expressing the hope that by 1980 in- 
juries would be reduced by ''50 percent or something like that. I have 
no basis for doing tliat other than hope." ^" 

Detecting OSHA's impact on injuries and illnesses would be a rela- 
tively simple matter if it were indeed as large as its most optimistic 
sponsors hoped it would be. Such expectations, however, were unreal- 
istically high. Unfortunately, the far modest gains in worker safety 
and health that might reasonably be hoped for are considerably more 
difficult to detect. The problems of measuring OSHA's impact on the 
incidence of occupational diseases are overwhelming. Most occupa- 
tional illnesses, particularly the more serious ones such as cancer, re- 
sult from the cumulative' effects of prolonged exposure to health 
hazards. Many of the cases of occupational disease observed today are 
the result of exposures many years ago. Correspondinglv, the eifects 
of current exposures may not be felt for many vears, in some cases for 
several decades. Thus even if OSHA is successful in reducing the ex- 
posure of workers to health hazards, the reduction in illness rates will 
occur only gradually. Given the difficulties involved in determining 
precisely which cases result from occupational exposures and which 

^r^.Iio^^^^^ •^^^^^^^* ^^ ^^^^ ^^^'^^ ^^ possible to obtain precise measures 
of OSHA's impact on occupational health. 

Although direct measurement of the effects of OSHA's actiWties 
on occupational illness rates is. at best, many vears in the future it 
may be possible to develop indirect indicators in the shorter run Dr 
Corn proposed that OSHA emplov the concept of "workers re- 



movea from risk." ^^^ That is, when a health hazard is abated as a re- 
sult of OSHA enforcement activities, the reduction in the number of 
workers exposed to that hazard would be recorded. Used in conjunc- 
tion with data on the risks of morbidity and mortality associated with 
the hazard, such information could then be used to estimate the ulti- 
mate impact on workers' health. Those estimates would be far from 
perfect, and there would be a considerable danger that inspectors 
would overstate their accomplishments, but they would provide at 
least some basis on which to assess OSHA's performance in the health 
area. No such basis exists now, although the paucity of health stand- 
ards and the minimal resources devoted to their enforcement would 
suggest little impact. 

In contrast to occupational illnesses, OSHA's impact on injuries and 
accidental deaths might reasonably be expected to be measurable in the 
short term. OSHA has been in existence now for 7 years and has de- 
voted the bulk of its efforts, both in promulgating standards and 
enforcing them, to the safety area. Furthermore, many of the prob- 
lems involved in measuring OSHA's effect on illnesses are not present 
with injuries. Reducing the exposure of workers to genuine safety 
hazards should have an immediate impact on the injury rate; unlike 
health hazards, the effects are not cumulative and there are no long 
latency periods. Finally, there are no problems with distinguishing 
between occupational and other injuries. 

Measurement of OSHA's impact on injuries is, however, compli- 
cated by a number of factors. The first results from the Act itself, 
which required a substantial alteration in the data collected on occu- 
pational injuries. Prior to the OSHAct, the Bureau of Labor Statistics 
collected data from firms on a voluntary basis, under the Z16.1 stand- 
ard. Since the passage of the Act, data have been collected on a more 
comprehensive basis, with mandatory reporting by firms and a broader 
definition of injuries. Thus data before and after the Act are not 
directly comparable. Unfortunately, the differences between the two 
sources of data cannot be reconciled simply by applying a uniform 
factor across industries to account for better reporting under OSHA. 
During the first year that the Act was effective, the Bureau of Labor 
Statistics attempted to conduct a survey from which conversion factors 
by industry could be constructed ; that project had to be abandoned 
because of inadequate funding and a low response rate.^^* 

Even had there been no changes in the procedures for classifying 
and collecting work-injury data, measurement of OSHA's impact 
would remain a difficult task. An assessment of the effect of OSHA on 
injury rates requires that one be able to specify what they would have 
been in its absence. But injury rates prior to OSHA were not stable, 
nor were they changing in a regular, easily predictable fashion. Table 
4 presents lost- workday injury rates per million hours worked re- 
corded under the old Z16.1 standard for manufacturing industries over 
the decade preceding passage of the OSHAct. Over that decade, the 
lost- workday frequency rate rose a total of 26.7 percent, for an average 
annual increase of 2.4 percent. The year-to-year changes, however, 
were far from uniform, ranging from —1.7 percent for 1960-61 to 

i°3 Interview with Corn. 

154 U.S. Senate, OSHAct Review 1974, pp. 968-971. 


-f 6.3 percent, for 1965-66. For the last pair of years during v/hich data 
were collected under the Z16.1 standard, 1969-70, the increase ^ras 2.7 
percent. The severity rate, a measure of workdays lost to injuries and 
accident deaths, rose only 0.8 percent from 1960 to 1970, but year-to- 
year fluctuations ranged from —7.3 percent for 1960-61 to 4-5.8 per- 
cent for 1968-69. Data for other industries show similar year-to-year 
















... 753 











Source: The President's Report on Occupational Safety and Health, May 1972, p 71. 

As of March 1978, data on injury rates collected under the new pro- 
cedures v.'ere available for five calendar years: 1972-76. For manu- 
facturing industries as a whole, the lost-workday rate, tlie one most 
comparable to the data in Table 4, rose 7.1 percent from 1972 to 1973 
and an additional 4.4 percent in 1974. It fell 4.3 percent in 1975, a re- 
cession year, but then rose 6.7 percent in 1976.^^^ The 1972-76 average 
annual increase was thus about 3.3 percent, slighty above the 1960-70 
pre-OSHA average. The overall injury and illness rate, which in- 
cludes injuries not resulting in a lost workday, for the entire private 
sector rose slightly in 1973 and 1976, but fell 5.5 percent in 1974 and 
12.5 percent in 1975, for an average annual decrease of about 4.2 per- 
cent. This decline, however, has been due entirely to a reduction in 
less serious injuries; over the same period the lost-workday rate for 
the private sector rose at an average annual rate of 1.5 percent. The 
most recent increase, from 1975 to 1976, was 6.1 percent. ^^"^ Little can 
be inferred about OSHA's effectiveness from the raw data, although 
they do make clear that its impact on injuries has not been dramatic. 

Evaluation of OSHA's net impact on accidental injuries and 
deaths requires tlie careful use of statistical procedures to con- 
trol for confounding factors. Several studies have been made 
which attempt to control for these other factors, primarily through 
the use of multiple regression techniques. Although the results of 
these studies are by no means conclusive, they provide the best evi- 
dence currently available. 

In a recently completed report prepared for the Department of 
Labor, John Mendeloff attempted to measure the impact of OSHA on 
the rate of chancre in the lost- workday injury rate in manufacturing.^^^ 
By using the change in the injury rate, rather than the rate itself, 
as the dependent variable, ^lendeloff was able to use both pre- and 
post-OSHA data. He fi.t a linear regression model employing four 
independent variables (the new hire rate, the new hire rate lagged 

i"'Data for 1972 and 107.''. contained in a T'.S. Department of Labor. BLS. press 
re'ease. Ttf^r. 20. 1074. Data for 1973 and 1074 contained in a release dated Nov. 19, 1975. 
Data for 1974 and 1975 contained in a release dated Dec. 8, 1976. Data for 1976 contained 
in ;i re]P!\se (hited Dec. 1. 1077. 

1^ Tbid. 

''* Mpiuleloflf. "An Evaluation of the OSHA Program's Effect on Workplace Injury 

83-944—78 15 


1 year, the percentage of male workers aged 18-24, and the average 
hourly wage of production workers) to injury data collected under 
the Z16.1 standard from 1948 to 1970. The new hire rate variables 
were included to control for the fact that inexperienced workers tend 
to have higher injury rates and because the new hire rate is closely 
correlated with the business cycle, which in turn has been shown to 
have an effect on injury rates. The variable for percentage of male 
workers aged 18-24 controlled for the fact that those workers have 
above-average injury rates. The wage variable was included to cap- 
ture the cost of injuries to firms. The model was then used to predict 
the 1972-73 and 1973-74 changes in the lost-workday rates recorded 
under the new procedures mandated by the OSHAct. The test of 
OSHA's effectiveness is then to compare the predicted changes with 
those actually observed ; if the observed changes are lower than pre- 
dicted, then there is some basis for arguing that OSHA has depressed 
injury rates below what they otherwise would have been. Mendeloff 
reports that the observed percentage changes in the post-OSHA 
injury rates were slightly lower than predicted by his model. For 
1972-73, the predicted increase was 7.9 percent, while the observed 
was 7.1 percent, a difference of 0.8 percentage points. For 1973-74, the 
difference was more substantial, 2.2 percentage points, with the model 
predicting a 6.6 percent increase as compared with the actual rise of 
4.4 percent. These results, however, are not statistically significant; 
the apparent decrease may only reflect random variations in the data. 
His results with the same model and several variants of it using Cali- 
fornia workmen's compensation data are similarly inconclusive; in 
some cases the observed rates for 1973 and 1974 were actually higher 
than predicted, which would suggest that OSHA has served to in- 
crease injury rates in California, a highly implausible result.^'^^ 

Mendeloff points out, quite rightly, that the tests described above 
are capable of detecting OSHA's impact only if it is large. In a series 
of additional tests he attempts to measure OSHA's impact on those 
types of injuries and fatalities which a panel of safety engineers in 
the California Division of Industrial Safety identified as most sus- 
ceptible to regulation through a system of standards enforced via 
inspections. Using California data, he finds that the observed rates 
for two types of preventable injuries — caught-in-and-between and 
falls and slips — appear to have been significantly lower than pre- 
dicted in 1974. He uses those results to estimate that OSHA reduced 
the injury rate in manufacturing by at least 2 percent to 3 percent 
in California in 1974.^^^ Also based on an examination of preventable 
accidents, he estimates that OSHA has reduced overall work fatalities 
in California by about 5 percent, with most of the reduction occurring 
in the construction industry.^^" The degree to Avhich these results may 
be generalized to the ?^ation as a whole is uncertain. California has 
opted to run its own State plan under section 18(b) of the Act, and 
had a relatively strong OSH program even before the Act's passage. 

One explanation for the apparently small impact of OSHA on 
overall work injury rates is that only a small proportion of covered 
firms were actually inspected during the first few years of OSHA's 

3"^" Ibid., pp. 45-61. 

^'^ Ibid., pp. 61-78, 105. 

i«eibid., pp. 70-84, 106. 


existence. In theory, the threat of being fined for violations uncov- 
ered during an insjpection provides an incentive for firms to comply 
with standards before being cited. In practice, as discussed earlier, 
for most firms the probability of being inspected is so low and the 
fines that are proposed for violations are so small that there is little 
incentive to come into compliance until citations have been issued. 
At that point, compliance becomes cost effective, both because the 
probability of reinspection goes up significantly and because the pen- 
alties for failing to abate hazards are substantial. 

If firms in fact do not comply with OSHA standards until they 
have been cited, it may be most appropriate to measure OSHA's im- 
pact on the injury rates of inspected firms. Two such studies have 
been made, one employing industry-level data, the other using data 
on individual firms. 

As described earlier, firms within the five industries selected for 
inclusion in OSHA's Target Industry Program (TIP) were subject 
to a much higher probability of inspection than other firms. Thus, if 
inspectors are effective in abating hazardous conditions, firms within 
tlie target industries should have experienced a greater reduction in 
injury rates than other firms. In order to test that hypothesis, Robert 
Smith estimated a model in which an industry's post-OSHA injury rate 
(he used rates from both 1972 and 1973) was written as a function 
of its injury rates in 1968, 1969, and 1970, the ratio of its employment 
in the post-OSHA year to that in 1970, and a dummy variable indi- 
cating whether or not the industry was included in the TIP.^*^^ losing 
both 1972 and 197:i national injury rates as dependent variables, the 
estimated coefficients for the TIP variable were positive, but statisti- 
cally insignificant. That is, contrary to expectations, injury rates in 
the TIP industries increased more, not less, than expected. Smith's 
results with a similar model employing California workmen's com- 
pensation data were qualitatively similar; TIP injury rates were 
slightly higher than predicted, but not significantly so. In a more 
recent paper. Smith reports additional results based on a slightly dif- 
ferent model which show the TIP exerting the expected effect on 
injury rates; the coefficients, however, remained statistically 

Aldona DiPietro, an economist on the staff of ASPER in the De- 
partment of Labor, has recently reported the results of her analysis 
of the effects of OSHA inspections using firm-specific data for 1972 
and 1973.*" By using data on individual firms, she is able to estimate 
the mipact of OSHA inspections within particular industries. Her 
estimating equation is similar to Smith's, expressing the injury rate 
of a firm in 1973 as a function of its injury rate in 1972, the propor- 
tional change in its level of employment, and whether or not it was 
inspected during 1973, either by a Federal or a State inspector. She 
estimated this model separately for each of 18 different two-dio-it SIC 
industries and, within each industry, for small (less than foO em- 
ployees), medium (100 to 500 employees), and large (more than 500 
employees) firms. Her results show that in most industrv-size groups, 
neither Fe deral nor State inspections have a significant impact on 

^^^:h^^^^ '' "'""'^ ^"-^ ^°^"^*- ^-- 

'•"^ Smith. The OSHAct, appendix C. 
Indus1rle?^m2-73'''^Dra\f ^^Tl'.hnfpni^^^ ^^^C" J"^P^<^U*L° Program In Manufacturing 


injury rates. More disturbingly, in those cases where the coefficients 
are significant, they are far more likely to be positive, suggesting that 
inspections increase the number of injuries in a firm. DiPietro dis- 
cusses several possible explanations for these unexpected results. First, 
she notes that inspections conducted in 1973 may not have had any 
impact during that year. She reports, however, that when the inspec- 
tion data are further broken down by the quarter in which the inspec- 
tion was conducted, she finds that earlier inspections have no more 
effect on the injury rate than later ones. Her second possible explana- 
tion is that firms may improve their recordkeeping following an in- 
spection. If that were the case, however, then firms inspected earlier 
in the year should have had the greatest increases in recorded injuries ; 
as reported above, there were no significant differences across the 
•course of the year. 

DiPietro's final and most plausible explanation of the apparently 
perverse effect of inspections is that inspectors have managed to 
identify firms with higher than average injury rates within each in- 
dustry and have concentrated their inspection efforts on those firms. 
Her estimates show that Federal inspectors are indeed more likely to 
inspect high injury rate firms, especially those firms whose injury rates 
rose from 1972 to 1973. State inspections appear to be less concentrated. 
Thus the causation may be reversed; inspections do not increase in- 
jury rates, but rather an increase in a firm's injury rate raises the 
probability that it will be inspected. Some of these difficulties may be 
resolved when 1974 firm-specific data become available, making it pos- 
sible to estimate the impact of inspections on injury rates a year later. 
Another problem may arise then, however: If the firms inspected in 
1973 included a disproportionate number with unusually high injury 
rates, we would expect those firms to have lower injury rates in 1974, 
independent of whether or not they were inspected.^^* 

The results of the studies summarized above, while far from con- 
clusive, suggest that OSHA's impact on injury rates has been minimal 
at best. It seems likely that OSHA's impact will grow somewhat over 
time and that that impact will become susceptible to more accurate 
measurement as more and better data become available. 


The broad impact of the OSHA on firms is suggested by the fact 
that 84 percent of the companies responding to a survey conducted by 
the National Association of Manufacturers reported that they had 
undertaken new safety and health efforts as a result of the act.^^^ The 
costs imposed on firms by the act range from the obvious and easily 
measured cost of fines paid for violations of OSHA standards, to more 
subtle and difficult-to-measure expenses, such as productivity losses 
due to OSHA-mandated changes in work practices. To the extent that 
compliance with OSHA's standards promotes a safer and more health- 
ful workplace, of course, firms may reap benefits in terms of reduced 
wage premiums for risk, lower turnover, reduced absenteeism, smaller 
insurance premiums, and the like. Most costs, of course, will be passed 
on to consumers in the form of higher prices. 

1"* An extreme example Illustrates this point Suppose the prohaUilitv of an aorident 
is 0.1 per year in e^irh firm and that inspectors visit only firms whicli have accidents 
Thus the inspected firms will have an injury rate of 1 and t'^e uninspected firms wiji 
have a rate of zero. Suppose inspections make no difference. If we examine the iniurv 
rates of firms 1 year later, we will find that the average for inspected firni'; will have 
fallen from 1 to 0.1. while the average for uninspected firms will have increased from zero 
to 0.1. malting it appear that inspections have a dramatic impact on injury rates. 

"5 Ashford, Crisis, p. 317. 


The direct financial burden imposed on firms by fines has been rela- 
tively small, totaling only $9.5 million in 1975.^^^ or about $2 for each 
of the approximately 5 million workplaces covered by the act. Even 
for the small percentage of firms inspected and fined, the average 
penalty per citation was only $163 in 1975.^^' It should also be stressed 
that by the usual precepts of benefit-cost analysis, these penalties are 
not real resource costs ; they merely represent a transfer from one seg- 
ment of the economy (cited firms) to another (the Federal Govern- 
ment) . 

In addition to the penalties assessed for violations, firms experience 
a variety of other expenses in connection with OSHA inspections. 
They must maintain the records of injuries and illnesses required to be 
shown to the inspectors. Durinjr the actual period of the inspection, 
management personnel must cease their normal duties in order to ac- 
company the inspector. 'WTiere the employer agrees to pay workers for 
time spent exercising their walkaround rights, that cost must also be 
borne by the firm. If the firm decides to contest any citations issued, 
additional expenses will be incurred in the appeals process. Although 
no reliable estimates arc available of these additional costs associated 
with inspections, it is probable that they far exceed the direct penal- 
ties paid. 

The major financial impact of the OSHA on firms results from 
complying with OSHA's safety and health standards, either on a 
voluntary basis or following citations. Some of these costs are direct 
and relatively easy to identify, such as the purchase and maintenance 
of specific safety equipment, or the modification of plr.nts to meet 
standards. However, even in those rases it may be diflicult to deter- 
mine exactly which oxpenditures should be charged against OSHA. 
For example, if an old piece of equipment is replaced with a new one 
which incorporates required safety features, it clearly would be inap- 
propriate to count the entire cost of the new equipment as a cost of 
OSHA. The appropriate amount would take account of the expenses 
of premature replacement and of the incremental costs of the OSHA- 
required features. 

In addition to expenditures on equipment and facilities, the costs of 
OSHA may include productivity losses associated with safety devices 
or witli safer work practices. For example, certain types of guards on 
woodworking machinery make it more time consuming to make adjust- 
ments and to replace blades. The inflationary impact statement on the 
coke oven emissions standards estimates that the work practices re- 
quired by the standards would reduce average productivity per worker 
by at least 18 percent, possibly by as much as 29 percent. ^^^ These pro- 
ductivity losses would be on top of substantial capital costs. 

Precise estimates of the total costs to firms of complying with OSHA 
standards are unavailable. The results of a survey conducted bv the 
National Association of Manufacturers in 1974 are shown in table 5. 
For firms in the smallest category, tliose with 100 employees or fewer, 
the average estimated cost per firm of complying with OSHA stand- 
ards was $35,000. For tlie largest f:]-iiis. those with over 5.000 em- 
ployees, estimated costs reached $4.7 million per firm. The reliability 
of these estimates, however, is open to question, as most of the firms 

'«« Job Safety and Health, 4 (May 1976), p. 3. 


I''' U.S. DOL, "IIS : Coke Oven Emissions," p. 81. 


responding to the survey had not actually made the expenditures, but 
rather were making rough guesses on the basis of limited information 
and experience. There is also, of course, an incentive for firms opposed 
to OSHA to inflate their estimates of compliance costs. 


Number of employees Safety Health Total 

1 to 100 24,000 n.OOO 35,000 

101 to 500 "]I""I""I 50,500 23,000 73,500 

501 to 1,000 ' ' . 141,140 209,627 350,767 

1,001 to 2 000 272,000 58,630 330,630 

2,001 to 5,000 r./.-I 552,000 278,000 830,000 

More than 5,000... -"..I 2,226,500 2,455,000 4,681.500 

Source: D.C. Kalis, "Update on the Cost of OSHA Compliance," Occupational Hazards (August 1974) p. 42. 

Dun's Review, a major business publication, has estimated that as 
a result of the OSHAct costs in many industries could rise by as much 
as 5 to 10 percent.^^^ Several industries are expected to have particu- 
larly high expenses, including producers of metal-stamping machines, 
where the 5 -year cost of complying with OSHA is put at $6 billion. 
A member of the National Association of Home Builders has predicted 
that OSHA requirements will add between $2,000 and $3,000 to the 
cost of the average new home.^^° 

Since 1973, McGraw-Hill has conducted a survey of safety-and- 
health-related capital investments. The resulting estimates overstate 
the costs of OSHA in that they include expenditures that would have 
been made even in its absence. Conversely, they understate OSHA- 
related costs because they measure only capital investments and not 
operating costs. Data from the survey on actual expenditures for 
1972-74 and on planned outlays for 1975, 1976, and 1978 are presented 
in table 6. From 1972 to 1974, capital expenditures on employee safety 
and health increased 68.1 percent in manufacturing industries and 
declined slightly in nonmanufacturing industries, for an overall rise 
of 22.5 percent. To place these figures in some perspective, they repre- 
sent approximately 3 percent of all capital spending by busiiiesses.^'^ 




Tiillions of dollars) 


























Sources: 1972-75, 1978, McGraw-Hill "Third Annual Survey of Investment in Employee Safety and Health," cited in 
IVIendeloff, "An Evaluation of the OSHA Program's Effect on Workplace Injury Rates," table B-1. 1976: Occupational 
Safety and Health Letter, June 8, 1976, p. 6. 

169 "The Crushing Cost of Safety," Dun's Review (January 1972), pp. 53-54, cited in 
Ashford, Crisis, pp. 319-320. 

170 Ibid. 

171 Ashford, "Crisis," p. 319. 


Estimates of the costs of complying with specific OSHA standards 
are provided by several "inflationary impact statements" commis- 
sioned by OSHA. Although the accuracy of those studies is open to 
question, they provide the most detailed, and probably the most reli- 
able, estimates available. Bolt, Beranek, and Newman, a consulting 
firm in noise control, has conducted two such studies of the costs of 
complying with proposed noise standards. Its most recent report esti- 
mates capital costs, spread out over 5 years, of $10.5 billion for a 90 
dBA standard and $18.5 billion for an 85 dBA standard.^^^ Annual 
maintenance costs are estimated to be 5 percent of capital expendi- 
tures. At the current rate of capital investment on safety and health 
related items, slightly over $3 billion per year, as shown in table 6, 
the 5 year capital costs of complying with the 85 dBxV standard would 
be roughly equal to all other safety-and-health related capital 

Although the noise standard surely involves the most significant 
costs of compliance for industry as a whole, several other standards 
also impose substantial costs on particular industries. For example, 
OSHA has estimated that as a result of its coke oven emission stand- 
ards, the steel industry will incur capital costs of between $451 mil- 
lion and $860 million and annual costs in tlie range $173 million to 
$1,150 million, depending on how strictly the standards are inter- 
preted.^^^ The estimate of total costs for the proposed inorganic ar- 
senic standard is $110.8 million per year.^^* 

Much of the concern over the impact of OSHA on firms has cen- 
tered around the problems faced by small businesses. Several factors 
have been cited which contribute to the special difficulties experienced 
by small firms in complying with the provisions of the Act : (1) Small 
firms often do not have personnel with the expertise needed to inter- 
pret standards and to make the necessary changes; (2) Small firms 
generally have a limited ability to raise the capital needed to make 
any major alterations required to comply with standards; (3) Econ- 
omies of scale may make it relatively more expensive for small firms 
to institute health and safety measures; and (4) Many small businesses 
are already marginal operations, without sufficient profit mai-gins to 
absorb the added costs of production associated with compliance. 

The language of the Act makes clear that its provisions are to apply 
to "each employer," without any exemptions based on size. The Act 
does include an amendment to the Small Business Act, providing for 
loans to aid small businesses in complying with the OSHAct, but rela- 
tively few loans have been made under this program. During the first 
5 years of its existence, only 156 firms secured loans, totaling $31.3 

A number of additional efforts have been made, both legislatively 
and administratively, to reduce the impact of the OSHAct on small 
businesses. In 1072, the Congress included in the fiscal year 1973 ap- 

1"- T'.olt. Beranok and Newman, Inc., "Economic Impact Analysis of Proposed Noise 
Control Regulation," p. 3-1. 

'^ U.S. DOL. "IIS : Coke Oven Emission," p. 2-3. 

1"* I\S. Department of I-abor. "Inflationary Impact Statement: Inorganic Arsenic." 
Occupational Safety and Health Administration (undated), p. 1-5. Three alternative 
exposure levels are considered in the IIS ; the figures cited in this report refer only 
to the most stringent, the proposed standard. 

i"^U.S. Department of Labor. "Occupational Safety and Health Administration's Im- 
pact on Small Business," (Occupational Safetv and Health Administration, Washington, 
D.C.. undated), p. 19. 


propriations bill for the Departments of Labor and HEW a provision 
exempting firms with 15 or fewer employees from compliance for 1 
year. The entire appropriations bill was vetoed by President Nixon, 
and the final version signed into law did not include the small busi- 
ness exemption. In 1974, the House attached to the appropriations bill 
an exemption for all businesses with 25 or fewer employees, but the 
provision was deleted in conference. The final bill did, however, ex- 
empt businesses with 10 or fewer employees from most of the record- 
keeping requirements.^^® By administrative action, OSHA had earlier 
provided such an exemption for firms with 7 or fewer employees.^^^ A 
rider to the 1977 appropriations act exempted farms with 10 or fewer 
employees from enforcement.^'^ 

Data on the actual financial impact of the OSHAct on small busi- 
nesses are unavailable. However, OSHA's inspection strategy has 
placed strong emphasis on reaching larger firms, thus limiting the im- 
pact on small businesses. For example, during fiscal year 1975 only 
55 percent of the firms inspected had 25 or fewer employees, although 
nationally about 90 percent of all firms are of that size.^'^ In addition, 
OSHA's procedures provide for penalty reductions for smaller firms. 
Firms with fewer than 20 employees receive a 10 percent reduction in 
penalty, while penalties for firms with 20-100 employees are reduced 
by 5 percent.^^^ It remains the case, however, that small firms do ex- 
perience higher fines per violation per employee. 

OSHA's inability under the Act to provide onsite consultation serv- 
ices has been cited as placing a special burden on small businesses. Sev- 
eral efforts to amend the law to permit consultation have failed. How- 
ever, most of the approved State plans now in operation do permit 
firms to ask that a State official visit the worksite, point out any viola- 
tions, and suggest ways in which the firm can come into compliance. 
In addition, OSHA currently has contracts with State agencies in 13 
States without approved plans to provide consul tation.^^^ Finally, 
OSHA will provide off site consultation, either by phone or in the office, 
and NIOSH has industrial hygienists available to advise on abate- 
ment of health hazards. Despite the many complaints about limited 
consulting services, the available programs have not been used 

InfMion and employment 

Inflation and unemployment have been the primary concerns of U.S. 
economic policymakers over the past several years. OSHA's overall 
impact on prices and emplovment is impossible to assess. Potential 
effects may be identified, but hard evidence is lacking and quantitative 
estimates are limited for the most part to inflationary impact studies on 
specific standards. 

Inflation — The causal mechanism by which OSHA's activities might 
drive up prices is readily seen: Compliance with safety and health 
standards raises the costs of production. The degree to which these 
higher costs are passed on to consumers in the form of higher prices 

i'76 Asbford. Crisis, p. 180. 

I'^TT.S. DOL. "OSHA's Impact on Small Business." p. 8. 

^'^'^ Mpmorandnm from M. J. Bolle. Contrrppsional Research Service. 

I'^TT.S. DOL. "OSHA's Impact on Small Business." app. C. 

ISO Ibid., p. 22. 

i« Ibid., p. 26. 

182 According to Ibid., only about 100 such visits are made each year. 


will depend on a variety of factors, including the market structure of 
the industry, and the price elasticities of supply and demand. To the 
extent that compliance with standards reduces wage premiums for 
hazardous employment and other costs to the firm associated with 
accidents and illnesses, production costs will be reduced. A successful 
effort to improve occupational safety and health could, conceivably, 
also reduce the growth rate of demand for medical services, which 
have experienced some of the most rapid rates of inflation. 

Estimates from two inflationary impart statements suggest that in- 
dividual OSHA standards have negligible effects on overall price lev- 
•els. although the impact in particular sectors may be significant. The 
inflationary impact statement on the proposed inorganic arsenic stand- 
ard, for example, estimates that the price of arsenic trioxide would 
rise between 10.4 and 25 percent, depending on whetlier or not the only 
U.S. producer of arsenic metal and arsenic trioxide is forced to close.^^^ 
It is also estimated that the price of arsenic-based wood preservatives 
would increase more than thi^eefold, resulting in a 5- to 20-percent in- 
crease in the price of treated wood.'^* The inflationary impact state- 
ment on the coke oven emissions standard places the direct price im- 
pact in the primary iron and steel manufacturing industry in the 
range 0.42 to 2.3 percent, I'osulting in an overall price rise in personal 
consumption expenditures of between 0.01 and 0.07 percent. ^^'^ It should 
be stressed that these would be one-time price increases, and not perma- 
nent additions to the rate of inflation. 

As part, of its responsibility for evnluating the inflationary impact 
statements prepared by OSTTA and other Federal agencies, the Coun- 
cil on Wage and Prioe Stnbility under President Ford chose n broader 
definition of inflation : 

In onr view, inflation is more than simply a matter of possible increases in con- 
ventional indices snch as the consumer price index (CPT). Rathor, to the extent 
that thp benefits of a proposed re,2:nlation exceed its costs, that regulation is anti- 
inflationarv. evep though the benefits may not be reflected in a lowered CPT. On 
the other hand, if the costs of a proposed regulation exceed the benefits, it is 

While many would quarrel with this definition of inflation, the basic 
point is a <rood one: evaluations of OSHA, or of any regulatory pro- 
gram, should focus not on its impact on prices per se, but rather on its 
costs and benefits in terms of real resources. 

Employment — The direction of OSTTA 's net impact on employ- 
ment, let alone its magnitude, is unknown. In some cases, enforcement 
of OSHA standards will lead to a loss of iobs; exposure limits may 
force greater automation: the costs of compliance may raise prices, 
decreasing demand for some goods and hence lowerinor emplovment in 
those industries: some firms will be forced ont of business entirely. 
Converselv, compliance mav require the purchase of new equipment, 
and OSTTA-ii\^ndaterl wo^'k practices may lower prorluctivitv, thus 
generating additional demands for labor. In a smoothly functioning 
economy, virtually all of these effects would cancel out. Prices and 

^^-T.S. DOL. "ITS : Tnoreanic Arsonlo," p. Vl-f). 

i-^ Ibid. 

185 U.S. DOL. "ITS : Ookp Ovpn Emissions." p. 110. 

'"^.T. C. Miller TIT. "Statpniont on Bohalf of tho ronnnll on Waire and Price Stahilifv 
Before the Ooonpational Safety and Health Administration. Washingrton. D.C., Exposure 
to Coke Oven Emissions : Proposed Standards." Docket No. H-017A. 


levels of demand and supply would adjust until a new full-employ- 
ment equilibrium was achieved. In the present U.S. economy, how- 
ever, where prices and wage rates are not completely flexible and 
workers are not perfectly mobile, either geographically or occupa- 
tionally, significant employment effects may persist over a period of 
years. The argument often made by supporters of OSH regulation, 
that workers in dangerous jobs have no safe alternatives other than 
unemployment, also implies that those same workers will surely have 
trouble finding new employment if their jobs are eliminated through 

The inflationary impact statements on the proposed standards for 
coke oven emissions and for exposure to inorganic arsenic illustrate 
how OSHA standards may cut either way on employment. In the case 
of inorganic arsenic, the IIS estimates that between 2,900 and 3,700 
jobs will be lost if the standard is imposed.^^^ The impact would be 
especially severe in the arsenical wood preservative industry where, it 
is estimated, employment would decline by 1,600 workers, about 29 
percent of the current production force. ^^^ In contrast, the coke oven 
IIS estimates that employment in coke oven departments would rise 
by 5,000 or more due to a decline in productivity of between 18 and 29 

Recommended Directions for Eeform 

Our introductory caveat should be repeated : This report is an over- 
view of OSHA, not a definitive study. Nevertheless, given the impor- 
tance of the occupational safety and health issue, and the controversial 
nature of OSHA itself, we feel compelled to go beyond a mere review 
of the literature, to formulate tentative recommendations. Ideally, 
these recommendations will be validated, or rejected, in the not- too- 
distant future on the basis of more extensive studies of OSHA. 

Two considerations strongly urge an early evaluation of OSHA 
(for which this overview could serve as a motivating input ).^^° 

First, under the direction of both Eula Bingham and her immediate 
predecessor, Morton Corn, OSHA has embarked on a number of ini- 
tiatives, many of which respond to perceived inadequacies of past 
performance. Several of the important actions taken under Dr. Bing- 
ham are discussed in the Postscript to this report. It is as important 
that these initiatives be recognized and applauded if they prove bene- 
ficial, as it is that they be curtailed if inefficient or counterproductive. 
An evaluation can show where past criticisms either were misguided 
or have been met. 

Second, and more important, OSHA is devoting significantly 
greater resources to health protection, a field where the justification 
for Government intervention is relatively strong. OSHA's potential 
for protecting: workers will be substantially enhanced as its involve- 
ment with health grows. It is also true, however, that its ability to 
impose significant, quite possibly unreasonable, costs on both industry 
and workers will also be much greater. These new departures warrant 
an assessment of OSHA's performance and its methods. 

i"U.S. DDL, "IIS : Inorganic Arsenic," p. VI-11. 

is^Tbid.. vv. VI-11. VI-17. 

"0 U.S. DOU "IIS : Coke Oven Emissions." pp. 25, 81. 

1^ The Interagency Task Force on Workplace Safety and Health, created by President 
Carter and discussed brioflv in the Postscript to this report, hopefully will provide the 
kind of thoronprh evaluation advocated. 


The importance of this review task, we believe, justifies what might 
appear to be the presumptive nature of some of the discussion that 
follows. Still, we caution that the proposals that follow reflect in- 
sights gained in a brief period of study. They are presented more to 
stimulate discussion and thinking than to serve as a blueprint for 
reform of the agency. Whatever the merits of these proposals, there 
can be no question that OSHA is at a critical stage, and that bene- 
ficial reform could yield great dividends. A key task before OSHA, 
Congress, the executive branch, and concerned analysts is to determine 
which reforms are likely to prove beneficial. 

osha's role 

A Complement to Incentive Mechanisms 

Our earlier review of decenti-alized incentive mechanisms for pro- 
moting occupational safety and health — the market and the govern- 
ment-involving mechanisms, workmen's compensation, tort law, and 
taxes — suggests that those approaches have strong theoretical merits. 
The encouragement and provision of appropriate incentives should 
be a major part of any overall governmental effort to generate appro- 
priate levels of safety and health in the workplace. 

The primary government-provided incentive system in place at 
present is workmen's compensation. Tf compensation laws were modi- 
fied to tie each firm's costs more closely to its own safety record — 
through greater use of experience rating and requirements that insur- 
ance policies include substantial deductibles and coinsurance rates — 
incentives for providing safer conditions could be increased signifi- 
cantlv. A tax system designed to account for externalities — costs borne 
by neither the worker nor his employer, but rather society at large — 
should also receive careful consideration and study, perhaps as an 
adjunct to existing workmen's compensation systems. 

Decentralized incentive approaches are perhaps less promising as 
a primary approach in the healtli ai-ea. Problems associated with iden- 
tifying cases of occupational illness and of connecting them with spe- 
cific periods of employment render doubtful the use of workmen's 
compensation. Similar problems exist with regard to taxes on illnesses, 
although taxes on hazardous conditions, or even better, levies on 
worker exposure to health threats, might be a most effective means to 
promote OSH. Such taxes would probably be no more difRcult to 
monitor and enforce than the current standards system. 

Although we believe that decentralized systems can and should ulti- 
mately play a much more significant role than they do now in improv- 
ing occupational safety and health, we recognize that major 
responsibilities will and should continue to rest with OSHA. This 
judgment is both political and analytic. Congress has charged OSHA 
with a mission. The enabling legislation for OSHA made it quite clear 
that Congress did not intend that occupational health and safety be 
fully relegated to a decentralized framework, relyinir whollv on mone- 
tary mcentivcs. Moreover, contracts for occupational risks "and waives 
are drawn on far from perfect markets. Some of the more significant 
imperfections can best be dealt with through some form of centralized 
regulation, with OSHA the logical agency to conduct such efforts. The 
danger, of course, is that OSHA, while attempting to help the worker 


through regulatory interventions, may actually impose such substantial 
economic costs that consumers, shareholders, taxpayers, and even the 
protected workers per se, will suffer a net loss.^^^ This is particularly 
likely if, as has been too frequently the case with past OSHA proce- 
dures, the regulatory intervention does not substantially enhance the 
safety and health of workers, or if it imposes costs that are well out of 
proportion to the gains achieved. 

A final word should be said about the major advantage of incentive 
systems relative to the standards approach that forms the backbone 
of OSHA's present efforts. Incentive systems avoid rigidities that are 
inevitable with direct regulation that prescribes particular means for 
promoting OSH. If a firm will be penalized for a poor OSH record, 
as it would be with a tax system or a workmen's compensation add 
on, the firm will manipulate all of the factors under its control in an 
attempt to enhance OSH levels ; it will not concentrate on those that 
can be readily inspected. This flexibility may be of particular import 
in industries where there are attractive opportunities, either tech- 
nological or organizational, to pursue cost-reducing OSH-promoting 

The primary thrust of our recommendations is that OSHA should 
seek to channel its interventions to areas where it can be demonstrated 
to yield measurable benefits. This will require not only a reorientation 
of its standards-and-inspection system, but also a substantially im- 
proved capability to collect and process data, and disseminate informa- 
tion. Among these information-gathering activities, estimating the 
costs of meeting various standards should receive high priority. 

Health Versics Safety 

OSHA has devoted an ina*ppropriately large fraction of its efforts 
to the promotion of occupational safety as opposed to occupational 
health. This conclusion emerges from an examination of the justifica- 
tions for Government intervention in the two areas, and from a review 
of the evidence on the magnitude of the two problems.^^^ The Agency 
has been safety oriented from its inception. Its inspection force has 
been heavily weighted toward safety specialists as opposed to industrial 
hygienists, who could conduct health inspections. The overwhelming 
anajority of OSHA standards relate to safety problems. Most signif- 
icant health problems are not addressed by standards. 

The situation is changing. Standards for other health risks, includ- 
ing lead, trichloroethylene, beryllium, sulfur dioxide, ammonia, tolu- 
ene, inorganic arsenic, various ketones, and noise among others, are 

1*1 The ultimate Impact on workers of OSH regulation will depend on how the costs 
of compliance are distributed. Firms may reduce risk premiums or curtail employment, 
thus harming workers directly. Costs may be passed on in the form of higher prices, which 
will affect workers in their roles as consumers. Even if costs come out of profits, workers 
will be affected both as taxpayers, since lost revenues from the corporate income tax 
will have to be replaced by other taxes, and as shareholders of companies. Although 
individual workers own little equity, employee pension funds have significant holdings. 
P. F. Drucker, The Unseen Revolution : How Pension Fund Socialism Came to America 
(Harper & Row, New York, 1976), estim'ates that such funds will hold over one-half of 
all U.S. equity by 1985. 

i»2 There is evidence that underreporting is a serious problem with regard to occupa- 
tional health losses. See the speech of Dr. Morton Corn before the Manpower and Housing 
Subcommittee of the House Committee on Government Operations, May 12. 1976. He 
observes : "Taking this problem of underreporting into account, current estimates of 
tlie number of workers suffering harm from workplace exposure to toxic substances or 
harmful physical agents nevertheless range up to 100,000 deaths each year." Even if 
this estimate is overstated by a factor of five, the magnitude of the occupational health 
problem in the United States would appear to hare been underestimated, at least by the 


currently in various stages of the review process. OSHA is making a 
concerted effort to improve the management of the standards-develop- 
ment process and, in particular, to find some better way of dealing with 
the vast range of potentially toxic substances than smiply exanimmg 
them one at a time. In a major 3-year effort, OSHA will recruit and 
train compliance officers to deal with occupational health problems. 

OSHA should be credited for focusing more strongly on problems 
of occupational health, where it clearly has a significant potential 
contribution to make. This is also an area where, as we have suggested 
earlier, the potential costs of misguided regulation are much higher. 
The social costs as well as the health-promoting benefits of OSHA's 
expanding health programs should be carefully monitored, by Con- 
gress, the Administration, and OSHA itself. 



The major steps that OSHA is undertaking to improve its present 
set of standards have already been described. Some of the most widely 
criticized standards, primarily those adopted uncritically from con- 
sensus organizations, have been deleted or revised significantly. Major 
efforts currently underAvay or recently completed, both within OSHA 
and by a Presidential task force, should lead to beneficial changes in 
a subs'tantial portion of the remaining consensus standards. The stated 
goal of these efforts is to eliminate the unnecessary standards and to 
revise the others to make them easier to understand, with a greater 
emphasis on specifying the performnnce desired, railier tlir-n requir- 
ing particular methods of control. These efforts deserve praise, al- 
though final judgment must be reserved until the actual changes 
can be reviewed. 

OSHA, partly in response to external prodding, has shown an in- 
creasing sensitivity to economic considerations in the setting of its 
standards. The inflationary impact statements, prepared in response 
to a recently extended Presidential order, have had the desirable effect 
of making OSHA weigh, or at least formally acknowledge, the costs 
of complying with the standards that it establishes. OSHA's an- 
nounced intention of continuing these studies is most welcome, par- 
ticularly if OSHA can develop an in-house capability to feed such 
information directly into the standards-setting process. 

Many would argue that OSHA should not consider costs, except of 
the most extreme sort, in setting standards. Human life and health, it 
is argued, are priceless and should not be traded off against dollars. 
Indeed, efforts to make such trade-offs probably violate the spirit, and 
quite possibly the letter, of the OSHAct. We reply that while these 
trade-offs are often distasteful, they are also inevitable. Both as in- 
dividuals and as a society, we constantly make implicit choices to 
undertake risks to life and limb in order to obtain other valued at- 
tributes. Resources used to promote OSH are not available for other 
purposes, including efforts to save lives or improve health in other 
arens, such as programs for hiirhway safety or improved air quality. 

Ideally, each standard would be set on the basis of a careful anal- 
ysis of the costs and benefits it generates. Unfortunately, analytic 
teclmiqups to impute dollar values to such attributes as good health 
and greater longevity are neither well developed nor widely accepted 


at present. It should be stressed, however, that whether difficult trade- 
offs involving resources and health are dealt with explicitly and ana- 
lytically, or are ignored, they are still made every time a decision is 
reached affecting human health. Where occupational health decisions 
are concerned, we believe that a stronger emphasis on analytic ap- 
proaches is needed. 

Given the difficulties, both analytic and political, of valuing lives 
and health, we believe that the most promising approach for setting 
standards is to employ cost-effectiveness analysis. For each standard 
proposed, several different levels should be considered, including the 
possibility of no standard at all. Dollar costs should be estimated as 
accurately as possible for each possible level. The benefits associated 
with each level should also be assessed, measured perhaps in such 
terms as the number of disabling injuries or illnesses prevented, not 
in dollars. In many cases, the available data will permit only the crud- 
est estimates encompassing wide ranges of uncertainty. Still, we would 
argue, if the data available are too weak to make even crude estimates 
of the benefits, they also are too weak to establish standards following 
any procedure. 

Once cost and effectiveness estimates are assembled, at least rough 
answers can be provided to the question : As a standard is tightened 
progressively, how much does it cost per additional life saved (or 
illness/injury prevented) to achieve the higher level of protection? ^^^ 
This procedure would not by any means determine the actual level 
chosen. The vei-y process, however, would force OSHA to examine the 
consequences of its standards more closely and would also provide 
useful information for interested parties, including other executive 
agencies and the Congress. Perhaps more importantly, it would high- 
light inconsistencies in different areas; it might show, for example, 
that at current levels of stringency one standard costs $5 million at 
the margin per expected life saved, while another could be tightened 
at a cost of only $5,000 per expected life, thereby yielding 1,000 times 
the OSH gain for its cost impositions. In such a case, by loosening the 
first standard and tightening the second, it would be possible both to 
increase longevity and to free resources for other uses. 

If costs are an important consideration in setting standards, the 
level of protection afforded will vary significantly across and even 
within industries. In general, the lower the cost of compliance, the 
more stringent the standard should be. For example, it is obviously 
more expensive for a metal stamping firm to come into compliance 
with OSHA noise standards than it is for an electronics firm to do so. 
OSHA should consider setting standards at different levels for differ- 

193 Typically, as a standard is tightened the cost of achieving a fixed increment of 
health protection rises, sometimes quite rapidly. Consider a proposed standard that will 
prevent 100 injuries at a cost of $200,000. If OSHA were adopting policies that im- 
plicitly valued injuries at $3,000, and if this level of standard were the only option, it 
would seem well justified, since it prevents an injury at an average cost of •"«2,000. But 
other levels may be feasible. Suppose that a slightly relaxed standard would cost onlv 
$150,000 and would prevent 05 injuries. In that case, the marginal cost of preventing the 
last five injuries is substantial. To be precise, it is ($200,000— $150. 000)/100—f)5) = 
$10,000 per injury. If that $50,000 could be directed to more beneficial purposes elsewhere, 
including achieving more for the promotion of OSH, the slightly relaxed standard should be 

The cost-effective setting of standards requires that the marginal cost of preventing 
an injury <or saving a life) by tightening a standard must be the same for all standards. 


ent industries, or at least allowini^ a limited number of industries with 
very hio^h costs relative to benefits to meet less stringent standards.^^* 

The costs of compliance will also vary significantly from firm to 
firm within individual industries. Clearly it is impractical to set 
standards on a firm-by-firm basis, although flexibility in setting abate- 
ment periods for cited firms may achieve that goal partially. Signifi- 
cant cost savings could be achieved through tlie use of "grandfather 
clauses" in those cases where the costs of compliance are much higher 
with older plants or equipment than they are with new ones. Assur- 
ances that new standards would generally exempt existing plants and 
equipment would reduce the problems that firms now encounter in try- 
ing to anticipate future OSHA standards when undertaking major 
capital expenditures. For example, PTarvard University is currently 
planning to construct some new laboratories. In addition to making 
sure that the laboratories will comply with existing OSHA require- 
ments, administrators must also try to guess what additional stand- 
ards will be imposed during the buildings' lifctimes.^^^ 

Eelaxntion of OSHA's hostility to the use of personal protective 
devices could also generate very significant i*esource savings with little 
or no sacrifice in worker health and safety. OSHA has correctly 
pointed out the danger that workers may not use their protective 
equipment. This argimient lends indirect support for incentives-based 
approaches to OSH, for with them firms will have strong inducements 
to get workers to keep protective equipment in place. 

Given the limited size of its compliance staff. OSHA is able to in- 
spect only a tiny fraction of the worlcplaces under its jurisdiction each 
year. At current rates, the average employer will see a Federal OSHA 
inspector roughly twice a century. Thus the effectiveness of OSHA 
enforcement strategies will l->e a critical factor in determining its de- 
gree of success in ini]:>roving OSH. 

The threat of fines for noncompliance has been an empty one; the 
average fines are so low relative to tlie costs of compliance that for most 
firms delaying action until after receiving a citation is rational. One 
possibility would be to raise fines significantly in order to make delay 
le?s attractive. It would seem, how^n-er, with the current low proli- 
ability of inspection, that fines would have to be increased many times 
oyer in order to obtain much impact. Current exix^rience with serious 
violations suggests tliat very high fines mav make inspectors less will- 
ing to cite violations. Large fines for relatively minor violations might 
appear capricious, fui-ther damaging OSIIA's image, both with busi- 
ness and the public. Thus a major increase in the size of fines levied 
appears to be undesirable. Small fines have little incentive value, and 
are perceiAed by many businessmen as further evidence of OSHA's 
petty harassment. Indeed, the fiscal vear 1077 Appropriations Vet 
eliminates small fines in most cases.^^^ 

™i-f5^T',™'""""''''!'i""*'.'''°"' Harvard University administrators. 


OSHA's general worst-first philosophy for allocating inspections 
is a reasonable one, although the particular plan now in effect suffers, 
several shortcomings, as discussed earlier. As more data become avail- 
able, it should become possible to target inspections more accurately. 
For example, firms could be selected for inspection on the basis of their 
individual injury records, as reported to the BLS. The danger of 
such a policy, however, would be that it would provide an incentive 
for firms to underreport. Future statistical studies, similar to DiPie- 
tro's work summarized earlier, might show differences in the effective- 
ness of inspections across industries, thus providing another basis for 
setting priorities. 

OSHA's new National Emphasis Program (NEP) for foundries, 
which is intended to serve as a model for similar efforts in other in- 
dustries, appears to incorporate a number of attractive features: in- 
dustry participation through the major trade association has been 
encouraged, apparently with some success; inspectors have received 
special training to enable them to detect the hazards likely to be found 
in foundries; and inspections will cover health as well as safety haz- 
ards. Perhaps most welcome, OSHA has announced its intention of 
incorporating a strong evaluative component in each of the NEP's.^^^ 
If these inspections, involving a much <?reater input of resources than 
usual, cannot be shown to have a significant impact on occupational 
injuries and illnesses, it will provide strong support for the view that 
OSHA's approach to promoting OSH through the promulgation and 
enforcement of standards is misguided. Any evaluation of the NEP's 
should also include a careful accounting of the costs involved. 

Uninspected -firms and industries 

Under present OSHA procedures, and under any reform that seems 
likely, a vast proportion of American workplaces will go uninspected. 
Indeed, given the worst-first approach for general inspections, most 
firms that will not be inspected can predict that fact. If proper cal- 
culations are made, the deterrent effect of OSHA inspections and fines 
will vanish. Feelings of anxiety or annoyance about potential inspec- 
tions should also dissipate. 

A variety of responses to this situation are possible. We might pro- 
pose that a small proportion of inspections be directed at firms or in- 
dustries with strong safety records, where except for deterrent effects^ 
inspections must be expected to have low productivity. But since the 
deterrent effects of potential fines are already minimal, it is hard to 
see how moving from a zero to, say, a .0001 probability of inspection 
would change matters. The major argument for maintaining some 
finite probability of inspection for every workplace would be its sym- 
bolic value: OSHA cares about all workers; no firm is immune to 
OSHA; your workplace may be inspected; no, OSHA is not picking 
just on the high-risk firms. Though we see the argument for appear- 
ances and equity, a more vital objective is that OSHA direct its re- 
sources in the most productive directions, and we would not urge that 
random, ver^^ low probability inspections be targeted on low priority 
firms. Should it be acknowledged that such inspections will not be 
made? Probably yes. The effects on deterrence will be trivial. More- 

4 (AprTl'lGTer p^^SS^""'^^^^'" Program : A View from the Top," Job Safety and Health, 


over, employers who know they will not be inspected should look on 
OSHA with diminished apprehension and animosity. 

The basic problem remains. How do we feel when a regulatory effort 
with mandatory standfirds is only imposed on a small portion of the 
population ? Four options are open : . 

(1) We could accept this as the natural outcome of a cost-effective 
application of OSHA's resources. ^ 

(2) OSHA could attempt to recruit sufficient inspectors to have 
some reasonable probability of inspecting all workplaces. 

(3) We could dramatically raise fines so that even a low probabil- 
ity of inspection would generate some deterrent effect. 

' (4) We could find some other way of promoting OSH in low-prior- 
ity firms. 

Option 2 would be extraordinarily expensive, and would run directly 
counter to our proposal to channel OSHA to its most productive tasks. 
Option 3 seoms unrealistic given tlie traditional relationship between 
fines and the magnitudes of offenses in the United States. Option 4 
presents the only realistic alternative to option 1, the present system. 

Without detailed study, we would be loathe to propose any new reg- 
ulatory effort to supplement a standards system for those firms that 
cannot be inspected. What follows is not a firm recommendation, but, 
we hope, a promising possible approach. 

Inspection and compliance could be made the responsibility of the 
employer. He would submit a brief annual form declaring that he is 
in compliance. Some special procedures might be developed to secure 
worker input to the firm-certification procedure. 

Clearly any system of this sort could turn into an administrative 
nightmare, with some 5 million establisliments subject to OSHA 
jurisdiction. No more than say 20 or 30 standards should be listed on 
the form. They should be selected as a function of their level of health 
or safety threat (as documented by statistical studies), their expected 
cost of abatement, and their frequency of occurence. Different indus- 
tries might have different forms. Clarity would be a critical element in 
presenting the standards. 

Such a system might restore a deterrent where none now exists, 
promote some degree of equal treatment across workplaces, force 
OSHA to identify those violations which are of highest priority, and 
create the appearance of loss harassment in the workplace. Falsifica- 
tion of information could be made subject to a new system of fine 
structures. It might also fall under existing statutes which impose 
criminal penalties for falsifying information on official Government 
documents. As part o,f any type of self-certification system, it would 
be necessary to work out abatement procedures for violations. Here 
too bureaucratic involvement should be minimized. Below some mini- 
mum number of violations, a firm might be given a fixed period, per- 
haps until the next regular filing, to abate a violation. 

Compliance with a voluntarv^ system of this sort might be much 
higher than with the present low-probability-of-inspection system: it 
could hardly be lower. Employers who persist in violations when the 
only threat is a small fine might be most hesitant to lie on Govern- 
ment forms bearing their signatures, particularly with criminal penal- 
ties as a tliroat. Obviously, there still would have to bo some low 
probability of inspection to determine whether tlie attested levels of 
compliance wore in fact mot. 

8:5-044—78 16 


Provision of information^ consultation 

In the promotion of occupational safety and health, the clearest and 
best justified role for the Government is as a provider of information. 
To a limited extent, OSHA and its accompanying agencies have 
played this role. They have set standards in a great variety of areas. 
In theory at least, they should have set on the basis of careful evalua- 
tion o,f the data available. 

OSHA's major shortcoming on information provision, we would 
argue, has been its basic strategy of standards, inspections, and fines. 
It is perceived, particularly by industry, as a policing agency, not a 
consulting agency. At present, Federal OSHA inspectors are not per- 
mitted to engage in onsite consultation. This prohibition may help 
maintain the legitimacy and deterrent capability of inspections, and 
may prevent unmanageable demands on manpower. However, for each 
of four reasons, we think it should be reviewed : 

(1) It enhances rather than plays down the adversary relationship 
between the Government and industry in OSH promotion. 

(2) It prevents the Government from providing some quite useful 

(3) Less helpful oflPsite consultation, including such procedures as 
the review of blueprints, is permitted already. 

(4) There is a consistency and equity issue since most State plans 
allow for onsite consultations. 

In some nonplan States, the Federal Government has contracted 
with State officials to provide onsite consultations. Quite simply, the 
Federal Government has a clear responsibility to provide information 
on ways of promoting OSH : onsite consultation seems a good way. If 
consultation services turn out to be expensive, a marginal cost charge 
would be appropriate. 

The Federal Government can provide information that will inform 
bargaining and discussion relevant to OSH between workers and em- 
ployers. '\'\niatever its regulatory format, the Government should col- 
lect and disseminate information on the types o,f conditions that are 
most threatening to safety and health. At the very least, this could lead 
to improvements in OSH in the vast majority of the w^orkplaces where 
inspectors have never ventured, and where the deterrent effects of 
OSHA are minimal. For large workplaces, more likely unionized, this 
information could provide a helpful input to collective bargaining on 
work practices and working conditions. This could lead to a push to 
exceed OSHA specifications in some areas, and possibly worker as well 
as employer pressure to lower unrealistic standards. 

Some observers have proposed that where an established group of 
workers' representatives exists, and where that group, say a national 
union, has sufficient resources to evaluate information on OSH. safety 
and health determination should be left to the bargaining process be- 
tween management and workers. ^^^ If this aproach were followed, even 
to a limited extent, the Government could play a useful role by pro- 

1^"' John Dunlop. formpr Secretary of Labor, arprues in "The Limits of Lejral Compul- 
sion." unpublished paper. Nov. 11. 1975. that "the actions of various regulatory ajrencies 
need to be brou^rht into greater harmony with collective bariraininsr.'' He sufrprests that 
Oi=JHA seek to encourage barfraininc between labor and manajrement over OSH con- 
ditions, possibly following the general pattern set by the Labor Department when it ful- 
filled its labor-protection responsibilities under the Urban Mass Transit Act by brinsing 
toeether union and transit representatives to negotiate 8-year labor-protection agreements 
when Federal funds were used to take over failing private transit systems. 


Tiding information to inform negotiations. The negotiation or collec- 
tive bargaining approach, in contrast to uniform national standards, 
could respond to local needs and preferences, as well as differing cap- 
abilities for reducing OSH risks. 

In sum, the Federal Government has a most useful role to play in- 
providing information about health and safety risks. Though respon- 
sibilities in this area are recognized in overview speeches or docu- 
ments about OSHA, OSHA's performance, and the allocation of its 
resources, suggest that information provision is considered a respon- 
sibility of secondary importance relative to setting and enforcing 
standards. ^^^ Given the extraordinary difficulties of any approach based 
on standards, and most significantly the fact that the overwhelming 
majority of workplaces never directly encounter the system, serious 
consideratioTi should be given to placing increased emphasis on Gov- 
ernment information provision. 


OSHA should have the capability to process and compile the quite 
considerable amount of data that it collects on a systematic basis. Un- 
fortunately, however, the Government's data base on occupational 
safety and health, as now collected, is not well suited to address many 
important issues. The problems fall into three categories : 

(1) Lack of consistency or parallelism : 

(2) Gaps in the data : and 

(3) Inattention to data needs.=*'<' 

Since 1972. the Bureau of Labor Statistics has compiled an annual 
survey of occupational injuries and illnesses. Its sample consists of 
200.000 establishments, with State samples providing an additional 
400.000. Firms with more than 100 employees are surveyed every year. 
Beyond its individual establishment data! the Bureau publishes'an an- 
nual bulletin summarizing injury and illness data by SIC code. 

OSHA's management information system processes reports on all 
inspections.-" I^nfortunately, establishments are identified with dif- 
ferent codes on the BLS and OSHA systems. It is therefore impossible 
to combine the two data sets to investigate such issues as: What has 
been the relationship between inspections, violations, and subsequent 
accident or illness rates? The BLS does not collect information either 
on occupations of workers who suffer health losses or on OSHA- 
monitored conditions. This makes it difficult to draw any firm infer- 
ences from BLS statistics about the relationships between conditions 
m the workplace and subsequent OSH losses.=02 Jt clearly would be 
highly desirable to code Bureau of Labor Statistics and'OSHA in- 

s hulrze MMis. I • Scttin? National Priorities- (lirookin-s Institution Washin4o V) P 
1!M,, note that Innh OSHA and tl.e Consumer I'rodnrt Safety Coiun V^^^^^^^^ 
Mm?n^'h1./?v,*^^^.^'"'^ '^"'^ '^^''' ^"/^'•"^^^tion provision service, and \heiranaTvtic capa 

U r:;]""^ ^^'%^ r.mjrress has cut those requests, placing greater emphases on enforceinent 
r.^ZL\ '^ "^eful analysis of the way present and prospective dati bases could he em- 
Ao^^'^. ^ X- '^T'^ OSHA's impact, see A. DiPietro. ''Data N>eds for t le " Eva uatinn of 
OSHA s Net Imp;ict." Olhce of Evaluation. Office of the Assistant Secretar^s^fo^rPo^^ 
Evaluation, and Resear.-h. T'.S. Department of Labor. An- 19 lOTi ^^"^^^^-^ for Po'icj. 

-■^Interview with Jack Katalinas. Director. Office of Manasement Data Sv^items OSHA 
^nS^^'l^^^-^^XS:;^. V^'f^;. ^'^^'^'' ^^^^^^- «^ Pro.?IS.i^S^sS^ 

^c<^.^ol{\TZTln^^^^^ '^''^ ^"^^^^'^ ''^"^^ additional information on 

The Bn e-/n nf T Tnr^f.H .• "^ • ^^'^ ^fi'^''^ ''"'"'^ "^' rerpiired to keer, uniform records. 
fVnn, ,x !^i • "'^ Statistics IS currently eniraced in efforts to compile specific statistics 

fioni ^Norkmen s compensation records in a varietv of States ^raubtics 


formation on a common basis. The need for consistency will grow if 
OSHA makes a conscientious effort to focus its attention on high 
priority areas, areas where it can be shown to have made a difference. 
In order to identif-y such areas, data on injuries and health losses and 
on inspections and violations will have to be merged. 

OSHA retains responsibility for overseeing State programs, and 
for assuring that their standards do not dip below Federal levels. At 
present, there is no requirement that States maintain data on inspec- 
tions and violations in the same form as OSHA; many do not. Con- 
sistent data bases would make it possible to draw inferences on rela- 
tive performance, presently a nearly impossible task. 

Data on injuries are fairly systematically generated m our system. 
Injuries are observable events. Moreover, they are linked through the 
workmen's compensation system to financial transactions. Finally, 
there are ongoing systems to collect such information. Data on occupa- 
tionally related health losses are more difficult to come by. Losses 
may not show up for many years, and there may be many contributing 
factors which can not be disentangled. It is unlikely that it will soon 
be possible to assemble data on occupational health losses at the time 
of occurrence. The best OSHA can do is to create a system where 
reasonably strong inferences can be drawn. 

There will be two key elements in a system to record occupational 
health losses : a tally of causal relationships, and an exposure index. 
Causal mechanisms relating occupational conditions to health loss 
must be inferred from substantial data sets, where available, and 
laboratory experimentation. Together, such studies can provide in- 
formation on what levels of exposure to what types of conditions will 
induce what sort of health losses. It is the responsibility of NIOSH to 
assess causal relationships. We would urge that emphasis be placed on 
the final implicit step in any such determination : Assess what mag- 
nitude of health loss will be associated with any particular level of 

An exposure index should be a running tally showing what type of 
workers (perhaps classified by age, sex, and health status) have been 
exposed to what intensities of health-threatening conditions. It is the 
objective of OSHA, of course, to reduce exposures, particularly to the 
more severe health threats, OSHA recognizes the need for output meas- 
ures relating to exposure to health threats.^^^ A data base on health 
exposure is likely to do more than merely document OSHA's perform- 
ance in dealing with health risks. The very fact of its tabulation, we 
suspect, would reduce health risks, by making employers aware that 
they were being monitored. Moreover, once concrete output measures 
have been defined, OSHA will likely focus more resources in the health 
area, and those resources will be directed more effectively. 

In accord with our belief that performance will in large part follow 
the available indicators of outputs and inputs, we have urged that 
systematic cost estimates be computed for meeting alternative levels 
of OSH standards. Even if OSHA's mandate does not permit cost to 
be an explicit consideration in setting standards, we believe that the 

203 Dr. Corn, in an Interview, told us of his plans to tally individuals removed from 
various levels of exposure. We prefer a direct exposure index, since we believe it will 
be hard to determine if 100 individuals are removed from risk in location A whether 
100 have taken their place in location B. Either index, however, would be welcome. 


mere fact that information on cost is generated will insure that it plays 
a more significant role. 

The general principle is clear. OSHA should be generating data on a 
systematic basis that would enable it to evaluate all aspects of its per- 
formance. This need will grow if OSHA follows the central recom- 
mendation of this report that it focus its energies where it can do the 
most for promoting OSH for any given level of economic cost. 

€apa'bility for policy analysis 

This brief study makes no claim to have reviewed all of the informa- 
tion that may be available on the performance of OSHA. At the very 
least, however, it has made a conscientious attempt to secure all of the 
major studies that are available. OSHA itself has not generated, at 
least for outside consumption, any studies that detail the major conse- 
quences of the operation of that Agency for OSH or for the use of 
economic resources. 

By this stage of its existence, OSHA should be well equipped to 
document the benefits it has generated for safety in the workplace. 
Health benefits are more difficult to establish, largely because of the 
time la^. Nevertheless, some indexes of exposure to health risks obvi- 
ously are needed. 

The economic costs entailed by OSHA would be more difficult to 
tally, particularly since they would necessarily rely on estimates made 
by thousands, perhaps millions of different workplaces. Still, it is a 
disappointment that the agency has made no efforts to estimate such 

OSHA has developed a management information system that regu- 
larly accumulates data on its inspections, violations, citations, penal- 
tiep, followup inspections, et cetera. Moreover, it systematically secures 
information on accidents through the Bureau of Labor Statistics. 
However, there appears to have been little attempt to date to assess 
the future consequences of various OSHA actions. 

At the beginning of 1976 a small (a few full time professionals) 
policy analysis operation was begun at OSHA.^o* Even at full opera- 
tion, this staff seems insufficient to conduct the type of policy reView 
operations needed to direct ?n agency of this size in profitable direc- 
tions. At present a significant propoi-tion of OSHA's analytic endeav- 
ors, for example its inflationary impact statements, is conducted under 
outside contract. This may be appropriate, but in the absence of an in- 
ternal policy staff capable at least of evaluating contractors' studies, it 
seems unlikelv that analytic studies can become a significant com- 
ponent of OSHA policymaking. 

There is a recognized need, both inside and outside of OSHA, to 
document the agency's accomplishments. Moreover, if the agency is to 
serve the American public, or even survive, it must eliminate those 
workplace interventions that do more to harass the employer than to 
promote the health and safety of the worker. It should" be able to 
identify which types of violations of which tvpes of standards in 
which types of industries are most likely to place workers in danger. 
The resources of the Agency at present are not directed so as to permit 
such determinations. 

Wash\n-7o^n^D C^'^Su^^ 00^19^^^ ^^^^^""^ Analysis and Intejrration Staff, OSIIA. 


A well developed policy analysis capabilitjr at OSHA should pro- 
vide two further classes of benefits. First, it should facilitate the 
Agency's day-to-day pursuit of its goals. At a time when its direction 
and emphases may be undergoing rapid evolution, the ability to assess 
what each branch of the Agency is accomplishing may prove most 
worthwhile. Second, it should provide the Agency with the capability 
to engage in long-run planning. The present standards-inspection- 
and-fine approach to occupational health and safety is only one of the 
many possible strategies that Federal regulatory agencies generally, 
and OSHA in particular, might pursue. With the present limited 
analytic capabilities within the Agency, it is unlikely that a sys- 
tematic long-range strategy can be formulated for assessing and 
experimenting with alternative approaches. OSHA should establish 
as a top priority the development of a policy analysis capability that 
would enable the Agency to formulate concrete objectives, detail per- 
formance on those objectives, evaluate alternative approaches to gov- 
ernmental promotion of health and safety, and develop long run plans 
for OSHA. (We are unhappy to report that OSHA's current iiead. 
Dr. Bingham, appears to have abandoned OSHA's fledgling policy 
analysis capability.) 

Division of responsibility between States and Federal Goverrument 

At present 23 States have chosen to exercise their right under the 
OSHAct to regain responsibility for regulation of occupational safety 
and health. In accordance with the Act, the Federal Government as- 
sumes 50 percent of the costs of operation for State plans, and 90 per- 
cent of any initial planning costs. Originally almost all States applied 
'for approval. Enthusiasm among the States now appears to be 

The motivation for decentralized responsibility for occupational 
safety and health is the traditional one within our heirarchical system 
of .nfovernment : Returning responsibility to the lowest feasible level 
achieves efficiency through decentralization as well as greater sensi- 
tivity to particular local conditions and preferences. I'he ar.o:ument 
for Focleral floor on OSHA standards is tAvofold: (1) Congress has 
concluded that there is a Federal responsibility to guarantee each 
worker a safe and healthful workplace ^^^ and (2) in the absence of 
such a floor there might be an incentive for States to engage in mutu- 
ally disadvantageous competition to lower standards.^^^ 

This study has made no attempt to investigate the consequences of 
the current mixture of State and Federal activities. Comparison 
studies would be hard to conduct at this time, since State plans do not 
provide OSHA with the same data collected on its own activities. How- 
ever, since matters of implementation — for example, frequency and 
targeting of inspections, propensity to cite violations, and magnitudes 
of fines — may affect so significantly the effectiveness of any plan, 
in ways at present not well understood, it would be extremely difficult 

"''^ Cornell, Noll, and Weingast. "Safety Regulation," conclude from an investljration of 
Information on budgetary requests jsnd approvals that Congress has been pushing for an 
Increased Federal role, while the Nixon and Ford administrations advocated greater 

2'"^ If all markets were perfectly competitive, a situation of this sort could not arise. 
But since, for example, wages do not automatically adjust to guarantee full employment. 
States might find it desirable to compete for jobs by lowering costs through lowered OSHA 


to determine if States met the test of providing plans ''at least as 
effective" as OSHA. 

The whole matter of State plans might merit a predominantly ad- 
ministratively oriented investigation to see whether the mixed system 
is generating problems. It is difficult to find parallel experience for 
regulatory procedures which are left predominantly to the State in 
some locales, yet are predominantly Federal in others. It is at least 
worth considering whether States should be encouraged more strongly 
to adopt their own plans, or whether, in fact, we should move in the 
opposite direction, discouraging all but the most enthusiastic States 
from takinjTf responsibility for ()SITA. The simplest moans to >^han£re 
incentives would be to change the cost-sharing formula, though some 
States would l)e unlikely to chancre tlieir present choice of status even 
if the cost-sharing percentages went from 50-50 to 90-10 or 10-90. 


A favored way to influence the objectives and performance of an 

agency is to change its position within cfovernment hier-irchies. It is 
sometimes proposed, for example, that OSHA be brought out of the 
Department of Labor and be made an independent regulatory agency. 
Such an agency, it is suggested, would have more control of its own 
budget and afl'aii'S. and would be less sul)ject to various poliriral p^-es- 
sures when initiating programs. To investigate the ramifications of a 
proposal of this sort would require a sensitive and lengthy political 
and bureaucratic analysis, and is thus well beyond the confines of this 
study. One point can be made: Any reorganization of this soit would 
require great energies. Even if gaining independent status would be 
desirable, on its own. it niav bo nndocimble if the required energies 
would be drained from other reform efforts. 

Further study should also be directed at the ro]ntirnQ]iips l)otween 
OSPTA and its cooperatinnr organisations. XIOSTT. OSTTRr and 
XACOSIT. ^lo>\ significantly, it is frequently alleged that XIOSTT 
belongs within OSIIA. At present, it is evident that XIOSTT i^ not 
providing OSHA all of the information it needs for setting standards 
on a timely basis. XIOSTT is certainly not equipped to investigate eco- 
nomic costs, or to relate standard setting to performance in the field. 
A major consideration in any investigation of the organization of 
Federrl efforts for OSII u\\'j\\i be the ap]n-opriate location of XIOSTT. 


OSHA was created to respond to wluU the TTon.'^e Committee on 
Education and Labor termed the "on-the-job health and safety 
crisis.'' -^' Congress hoped that by establisliing a Federal agency \o 
regulate OSH conditions it could ''reduce the number and severity of 
work-related injuries and illnesses." ^^^ It is now evident, after 7 years, 
that no dramatic iniprovonients have been achieved. Indeed, to date 
tliere has been no solid documentation that OSHA has yielded any 
gains in safety or health. This result is less di-couraging than it 
sounds. Data are inadequate: small gains would be difficult to docu- 

^ House Education and Labor Committee report on the OSHAct, reprinted In BNA, 
Job Safety and Health Act of 1970, p. 152. 

^^r.S. Senate Committee on Labor and Public Welfare. Legislative Historv of the 
OSHAct, p. 141. 


ment. Dramatic gains should not have been expected even under the 
most favorable circumstances. Only a small fraction of America's 
workplaces have been subject to inspection ; deterrent effects for un- 
inspected workplaces have been insignificant, given the low scale for 
fines. Moreover, OSHA has focused its attention on capital equipment, 
having virtually no control over such factors as worker or supervisory 
behavior, which contribute significantly to OSH. 

Any initial expectations that OSHA would make major inroads 
against occupational injuries and illnesses were totally unrealistic. 
Fortunately for the present image of the Agency, its OSH outputs are 
difficult to measure. Strong advocates of OSH regulation can be some- 
what assuaged by monitoring the Agency's input levels, particularly 
since they are likely to overestimate the returns from such inputs. 

The hostility of the business community, and indeed even of some 
labor representatives, to OSHA would have been more difficult to 
predict. Most employers in the United States have never met an OSHA 
inspector. Few have had to pay any significant fines, or incurred major 
expenses for correcting violations. It seems that OSHA's bad reputa- 
tion has traveled through chains of anecdotes. Uninterpretable or 
misguided standards, fines for conditions only most indirectly related 
to safety or health, and an occasional preposterous and expensive im- 
position to correct a violation have all been cited in both trade jour- 
nals and the popular press. The low likelihood of inspection has con- 
tributed to feelings that the whole system is somewhat random and 
therefore unjust. 

This situation has evolved at the same time that the public has be- 
come increasingly hostile to Government intervention into the econ- 
omy and into its daily lives.^^^ OSHA almost serves as a paradigm in 
this regard. The Agency is in an unenviable political situation, gener- 
ating frequent and bitter complaints yet yielding relatively little of 
the benefits its supporters expected. 

Rather than continue on the course of its first 7 years, we would 
argue, OSHA should be disbanded. Safety and health in the workplace 
would not suffer measurably, significant private and governmental 
resources would be saved, and an agency perceived primarily as a tool 
of Government harassment would be eliminated. Fortunately, these 
are not the only two courses available, and we would urge that some 
basic changes be considered. 

OSHA, and other Federal agencies promoting OSH. should channel 
their resources in three directions : (1) Generating, gathering, and dis- 
seminating information about conditions that promote OSH; (2) 
increasing the use of OSH-promoting incentive mechanisms as an al- 
ternative or complement to direct regulation; and (3) intervening 
in the market directly in those areas, and only in those areas, where 
there is a demonstrated relationship between the means of OSHA's 
intervention and safety and health. Regulatory procedures and stand- 
ards which can not be shown to be linked to occupational safety and 
health should be written off the books. Finally, we would urge that 

209 111 a series of lectnrps entitled "The Public Use of Private Incentives" CGodlcin 
lectures. Harvard University, Nov. 29-Dec. 7, 1976) Charles Schultze. now Chairman 
of the Council of Economic Advisers, noted this growing: dissatisfaction with Government 
recnlatinn. He argued that in many areas, quite possibly Including occupational safety 
and health, the Government should abandon the "command and control mode" of direct 
regulation, employing instead various Incentive mechanisms. 


OSHA be required to generate information systematically on the costs 
and benefits of its regulatory interventions, in that way guaranteeing 
that such information receives attention in political and administra- 
tive proceedings. OSHA now prepares inflationary impact statements 
under executive order, but legislation by Congress requiring that re- 
source costs be computed would provide a more direct, permanent, 
and legitimate means to introduce this consideration. 

OSHA, responding in large part to its precarious political posi- 
tion, is making many moves in appropriate directions. It is increas- 
ing the relative emphasis it places on health protection, where 
the risks are more difficult to recognize and externalities are greater. 
It is experimenting with more simply written and more performance- 
based standards. It is targeting its general schedule inspections 
in a more svstematic manner. These initiatives are welcome. But 
no agency with a commitment to its field, mr.cli less one as important 
as OSHA, can be expected to proceed full steam toward eliminating 
its marginal activities. Congress mu=t recognize that OSHA was never 
equipped for nor appropriate to the tasks for which it was originally 
conceived: it could hardly have succeeded. A change in em.phasis nnd 
direction is needed. Congress should releofislnte OSHA'<= mandate, 
requiring the Agency to pay explicit attention both to the health bene- 
fits that its interventions generate and to the resource costs that they 
entail. Such a requirement would encourage OSHA to rntionalize its 
interventions, directing them to the areas where they would be most 



The main body of this report was completed just as the appoint- 
ment of Dr. Eula Bingham as head of OSHA was announced in 
February 1977. Over the past year OSHA has undertaken a number 
of significant initiatives, the main thrust of which has been to in- 
crease OSHA's emphasis on health relative to safety. With consider- 
able fanfare, OSHA has announced its intention to rescind nitpicking 
regulations, and to concentrate on the serious threats to worker safety 
and health. In a widely quoted statement. Secretary of Labor F. Bay 
Marshall criticized OSHA's earlier approach as "chasing minnows 
and letting the whales get away." ^lo 

OSHA'S recent actions have generated a good deal of favorable 
publicity for the Agencv and its head. Dr. Bingham. The title of a 
recent OSHA pamphlet, "The Shift to Common Sense Priorities." 
summarize the new image that the Agencv is trving, quite success- 
fully, to create. In principle, we applaud the efforts of OSHA's new 
management. Few would argue that OSHA should noi eliminate the 
nuisance standards that antagonize business and do virtuallv nothing 
for worker safety. And, as we argue in the body of this report, the 
potential benefits from a greater emphasis on health hazards are sig- 
nificant. Health, however, is also an area where misguided regulation 
can prove exceedingly costly, swamping the health improvements 
achieved. Unfortunately, the evidence to date suggests that OSHA is 


not following the careful and selective approach to health hazards 
needed to secure major gains in worker health without imposing 
unreasonable costs. 

In this postscript we comment on two major initiatives of the past 
year: (1) the elimination of nuisance standards, primarily in the 
safety area, and (2) the increased emphasis on health, with special 
reference to the proposed new benzene standard and the proposed 
"generic" approach to regulating chemical carcinogens. We also dis- 
cuss, briefly, the work of the "Interagency Task Force on Workplace 
Safety and Health," created by Presidential order to examine ways in 
whicli Federal regulation of OSH could be improved, including the 
possible use of mechanisms other than standards and inspections. 


On December 5, 1977, OSH A announced its intention to revoke 
"more than 1,100 job safety rules that have no direct or immediate 
effect on worker safety or health." ^^^ The vast majority of the rules 
proposed for revocation were among the thousands adopted so hastily 
in 1971 from pre-existing consensus standards. They include some of 
the most widely ridiculed rules, such as toilet seat specifications and 
mounting heights for fire extinguishers. Also scheduled for revoca- 
tion are 10 of the 12 pages of the now infamous regulations on port- 
able wooden ladders. Until these standards are finally revoked, 
OSHA will treat violations of most of them as "de minimis;" that is, 
inspectors will notify employers orally, but will not issue written 
citations or propose penalties. Dr. Bingham has announced that these 
revocations are but the first step in a plan to "streamline and simplify 
job safety and health requirements." ^^^ Major revisions are planned 
for a variety of general industry standards, including walking and 
working surfaces, fire protection, and machinery and machine 

The elimination of nuisance standards is long overdue. Most never 
should have been adopted. The gains from eliminating these stand- 
ard'^. however, are likely to be realized primarily in tenns of a better 
public image for OSHA and a lessening of business antagonism, 
rather than in terms of significant gains in OSH or reductions in cost. 
Although the nuisance standards have infuriated many employers 
and have provided a rich source of amusing or outrageous anecdotes 
for OSHA's opponents, there is little reason to believe that compli- 
ance with them has been very expensive. We believe that OSHA's 
critics, particularly those writing in the popular press, have greatly 
overemphasized the importance of the nuisance standards, in part 
because these standards are so obviously misguided. We fear that as a 
result, some people will believe that the "OSHA problem" has been 
solved now that most of those standards are on the way out. Such 
optimism is r.nwarranted ; OSHA's more fundamental problems, par- 
ticularly its refusal to consider the economic costs of its actions in all 
hnt the most extreme cases, remain. 

211 See U.S. Department of Labor. OSHA. press release 77-1049, dated December 5. 1977. 

212 Ibid. 



Under Eula Bingham, OSHA is placing increasing emphasis on 
health hazards relative to safety problems, a process begun under her 
predecessor, Morton Corn. OSHA continues to expand its staff of 
industrial hygienists, needed to inspect for health hazards, and is 
emphasizing health threats in fne development of new standards. 
This increased attention to health is consistent with the past recom- 
mendations of most observers, including the authors of this report. 
We argue here that the magnitude of the occupational health prob- 
lem probably far exceeds that of occupational safety, and the 
arguments for intervention, based on poor information and exter- 
nalities, are considerably stronger for health. Occupational health 
problems, however, are usually complex and fraught with great 
uncertainties, and in numerous cases are very expensive to remedy. 
If regulations are to yield significant health gains without imposing 
excessive burdens, they must be designed carefully, with due con- 
sideration given to both protection and cost. In the past, OSHA's 
refusal to balance protection with cost to even the slightest extent was 
disturbing, but not of great consequence given the small number of 
standards it issued. If, however, OSHA is successful in its effort to 
speed up the issuance of healtli standards, the economic costs of ill- 
considered regulation could be stagirering, without a commensurate 
reduction in serioim liealth efl'ert-^. AVe discuss two specific examples 
in this regard: (1) the proposed benzene standard and (2) the pro- 
posed generic approach to the regulation of carcinogens in the 


In May 1077. OSHA issued an emergency tcMuporary stand- 
ard (lator stayed by court order) covering l>enzene. which lowered 
the permissible 8-hour time-weio-hted exposure limit from 10 parts per 
million fppm) to 1 ppm.'-^" Sliortly thereafter, the agency proposed a 
new permanent standard, that inchided a variety of requirements for 
medical surveillance and exposure measurements, in addition to the 
1 ppm oxpo'^ni-e limit. -^'' 

Benzene has long been reeof]:nized as a hazardous substance. Very 
high levels of exposure result in severe acute effects, in some cases in 
death. For at least 40 years, benzene has also been suspected of causing 
leukemia. There is little doubt that long-term exposure to benzene 
at high levels (in excess of 100 ppm, i^w times the existing 10 ppm 
limit) raises the risk of leukemia. The critical question, about which 
there is much dispute, is whether low levels of exposure pose a sig- 
nificant threat to health. 

OSHA's sense of great urgency, which led it to issue an emergency 
temporary standard, vras based on the results of a NIOSH study of 
workers in two Goodyear plants that produce clear plastic films using 
benzene.^i"^ xV follow-up study of the health status of workers who l>egan 
their employment at either of the plants prior to 1950 revealed a leuke- 
mia incidence rate several times higher than normal. OSHA has relied 

-'i-Tpilprnl Kopister. vol. 42, No. S.". (Mav 3. 1977). pp. 22516-22529. 
114 FofK'ral RedstPr. vol. 42. No. 10.'', fMav 27. 1977). pp. 274r)2-2747S. 
■''"• V.Y. Infante et al. "Leukemia Among Workers Exposed to Benzene," National Institute 
for Occupational Safety and Health, Cincinnati, Ohio, April 28, 1977. 


heavily on this study to support its claim that current occupational 
exposure to benzene are excessively high. Critics of the study, who are 
numerous, point out that little is known about the actual exposures 
workers in the two plants faced, and that what little evidence there is 
suggests that exposures were often far in excess of 100 ppm, sometimes 
as high as 1,000 ppm. Thus, critics charge, the NIOSH study adds 
little to our knowledge of the health effects of benzene exposure at the 
levels now found in U.S. workplace.^^^ Particularly suggestive in this 
regard is the fact that although the follow-up period extended to 
1975, no leukemia deaths in the cohort, were observed after 1961. 

OSHA's position on benzene, as it was on coke oven emissions, 
is that no safe level of exposure can be established for carcinoorens, 
hence exposure limits must be set as low as "feasible." "Feasibility" 
is established by showing that techniques are available to achieve the 
required level, and that the costs are not so high that significant num- 
bers of firms will be driven out of business. OSHA has made no at- 
tempt to estimate the health effects of the proposed standard, even in 
the crude fashion it did with the coke oven emissions standard. Under 
contract to OSHA, the Arthur D. Little Company estimated that the 
benzene standard would require $267 million in capital investment, 
with operating costs of $124 million in tlie first year and $74 million 
in subsequent years.^^^ Nowhere does OSHA indicate whether it be- 
lieves these hundreds of millions of dollars will save one life, a dozen 
lives, or a hundred lives per year. The contractor's report, which is 
over 100 pages long, with 200 pages of appendices, contains a grand 
total of 8 pages on the health benefits of the standard, 3 pages of which 
are simply a list of references. Moreover, most of the studies cited deal 
with the acute effects of exposure to benzene at levels far above either 
the current standard or the proposed one, and hence tell us nothing 
about the incremental benefits to be derived from the lower limit.^^^ 

The proposed benzene standard exhibits two additional disturbing 
characteristics. First, it continuos OSHA's opposition to the use of 
personal protection devices, in this case respirators, except as a tem- 
porary measure or where engineering controls and work practices are 
not sufficient to meet the exposure limit. As u'^ual, this refusal to per- 
mit the use of personal protection devices results in much higher costs. 
Second, the standard includes very detailed and extensive require- 
ments for medical testing and exnosure measurements. In most cases 
these tests and measurements niust be made quarterly, in other cases 
monthly. Fully two-thirds of the estimated first-year operating costs 
are due to these requirements; only one-third are for activities de- 
signed to lower exposure levels.^^^ It appears that these requirements 
were established arbitrarily, without considering a range of alterna- 
tives. For example, firms using benzene in only minute quantities, with 
concentrations below the limits of detection equipment, must ad- 
minister the same tests at the saine intervals as plants that just barely 
comply with the exposure limit. 

^K* These criticisms, made at the OSHA hearinjrs on Benzene in July and Au,iust. 1077, 
are summarized in E. W. Warren et al. "Post-Hearing Brief of API, NPRA and Individual 
Companies." Kirkland, Ellis & Rowe. Wasbin^ton, D.C.. September 22. 1977. 

21" Arthur D. Little, Inc., "Technology Assessment and Economic Impact Study of an 
OSHA Regulation for Benzene," draft report prepared for U.S. Department of Labor, OSHA, 

218 Ibid., ch. .?. 

218 Estimated first-year costs are $124 million, of -which $40.6 million would go for ex- 
posure measurements, $41,6 million for medical surveillance. Ibid., p. 1-9. 


Generic regulation of carclnogeTis 

In an effort to speed up the issuance of health standards. OSHA 
has proposed a generic approach to the regulation of carcinogens. 
Under the plan, proposed in October 1977, substances would be classi- 
fied into one of four categories based on evidence of their carcino- 
genicity.22° Substances for which th.ere is evidence of carcinogenicity in 
humans or in two or more animal tests would be placed in Category 
I. Classification in Category T would immediately trigger the issuance 
of an emergency temporary standard (ETS) and initiation of the 
steps needed to promulgate a pennanent standard. The proposal in- 
cludes model standards for both the ETS and the permanent standard. 
The latter closely parallels the proposed permanent benzene standard. 
Specific exposure limits are not included in the model standard, but 
rather would be set to the lowest "fensible" level (which could be no 
exposure if substitutes are available) for the particular substance in- 
volved. The model requirements for medical surveillance and exposure 
measurements are also similar to the proposed benzene standard, al- 
thouorh the specific medical tests to be required and the frequency with 
which they must be administered are left unspecified. 

Substances for which the evidence of carcinogenicity is only "suir- 
gestive." not sufficient to warrant jdncomont in Category T, would bo 
classified in Category TT. The model standard for Category TT sub- 
stances does not require that the exposure limit be as low as feasible; 
if a limit already exists and is considered adequate, it will remain 
unchanged. The model standard does, however, include all of the 
Cateirorv T requirements for medical surveillance and exposure 

Category TTT would be composed of substances foi- wh.ich there is no 
-evidence of '^arcinogcMiicity. Such substances would be listed and would 
not be subject to any regulation other than that required to protect 
against other, non-carcinogenic health effects. (The only substance 
listed thus far is ''water.") The final category, IV, is reserved for 
substances not used in I^S. workplaces, that may or may not bo 

Tlie concept of a generic a])proach to regulating hazardous sub- 
stances is an appealing one. XTOSH estimates that tliere are as many 
as 2.000 possible carcinogens in U.S. workplaf^es. T'nder OSHA's cur- 
rent substance-by-substance approach, there is little hope that a sig- 
nificant number of these substances could be regulated in the 
forseeable future. By lookins: at a number of sijnilar substances 
simultaneously, tradeoffs <^an be identified and made more consistent, 
and scientifir^ evidence and cost data developed for one substance can 
be brought to bear on others. Both tl^.e qu.ality of standards and tlie 
speed with which they can be pronmlgated could be increased. Un- 
fortunately, the current proposal for generic standards reveals little, 
if any, concern with improved decision making. The sole objective 
appears to be to speed up the process of issuing standards. 

Tlie sacrifice of quality to speed is revealed by two aspects of the 
OSTTA proposal: (1) the sole criterion for categorizing substances 
is the number and ty]x^ of tests showing carcinogenicity, and (2) 
tliere are only tv\'o categories for potential carcinogens to encompass 

"" Federal Reeister. vol. 42. no. 102 (October 4. 1977). pp. 54148-54247. 


about 2,000 different substances. In its rush to regulate, OSHA has 
ignored a wide range of vital distinctions. 

The classification system totally ignores evidence of the potency of 
potential carcinogens. Much controversy surrounds the choice of an 
extrapolative model to predict risk at low exposures based ouNdata 
obtained at high exposures. The controversy becomes still greater when 
the extrapolation must go from one form of exposure (say ingestion) 
to another (usually inhalation), or from laboratory animals to hu- 
mans. It is clear, however, that carcinogens do differ in their potency, 
and that any reasonable regulatory policy must recognize and deal 
with such differences. Where possible, crude close-response relation- 
ships ought to be developed, xlt the very least, some rough distinctions 
should be made on the basis of potency. Surely a substance shown to be 
carcinogenic to humans at exposures approaching current occupational 
levels ought to be treated differently than one for which the evidence 
is restricted to very high doses administered to animals by a route 
other than those encountered in the workplace. 

Of the 2,000 "suspect carcinogens" identified by NIOSH, probably 
more than 100 would be placed in Category I. an unknown, but pre- 
sumably small, number would be placed in Category IV (substances 
not used in U.S. workplaces), and the remainder would be placed in 
Category II. By adopting the goal of setting Category I exposure 
limits "as low as feasible," OSHA has continued its policy of ignoring 
costs in all but the most extreme cases, where limited economic analysis 
has been required by executive order. In the past, and we have little 
reason to expect the future to be different, OSHA has interpreted 
feasibility primarily in terms of technology, with little regard for 
costs ; that is, does the technology exist, or can it be developed, at what- 
ever cost, to meet the standard? Were we to take every technically 
feasible measure to improve health, we could easily exhaust the entire 
gross national product. In many areas we tolerate positive, often sub- 
stantial levels of risk, despite the availability of technologies or 
behaviors to reduce them. (Virtually all traffic fatalities could be 
eliminated, for example, by the technically simple expedient of lower- 
ing speed limits to 10 miles per hour.) We have decided, at least 
implicitl3\ that the costs of further reducing these risks outweigh the 
benefits. The OSHA concept of feasibility, embodied in the proposed 
generic standards, denies the legitimacy of even considering such 

As noted earlier, the proposed model standards for both Category I 
and II substances require extensive exposure monitoring and medical 
surveillance. Although cost estimates are unavailable, the evidence 
cited earlier from the proposed benzene standard, which contains very 
similar provisions, suggests that these requirements will prove very 
expensive : in the case of benzene, these requirements account for two- 
thirds of the first-year costs of compliance. As with the exposure lim- 
its, OSHA needs to adopt a more flexible approach that recognizes 
important differences across substances. For example, the frequency of 
monitoring and medical testing ought to vary with the potency of the 
carcinogen, the likelihood that exposures would vary over time given 
the production processes involved, the costs of the specific tests re- 
quired, and the nature of the health effects expected. In the case of 
Category II substances, where a major purpose of these requirements 


would be to gather data relevant to future reo:ulation. sampling pro- 
cedures could easily be devised that would reduce costs substantially, 
yet still provide high-quality data. 

In addition to the exposure monitoring and medical surveillancf 
provisions, the Category I and II model standards share a variety ot 
other requirements, including changing rooms and lunch rooms, man;/ 
of which will be costly. Thus the distinction between the two cate- 
gories, in terms of cost, will probably be relatively small and major 
cost burdens will be placed on the producers and consumers of as many 
as 2,000 substances. 

OSHA has had little experience thus far with health regulation, 
having devoted most of its resources to safety during its first 7 years. 
We are not aware of any evidence on the effectiveness of its existing 
health standards in reducing risk. OSHA has learned in the safety 
area, to its sorrow, of the dangers of overly hasty action. Now is not 
the time for OSHA to lock itself into a rigid framework for making 
regulatory decisions in the critical area of occupational health. The 
risks of precipitous action are simply too great, not just in terms of 
cost, but also in terms of workers health, since the unfocused ap- 
proach of the current proposal will dissipate scarce resources, includ- 
ing OSHA's, leading inevitably to the neglect of truly serious hazards. 


The final recent undertaking of note is rhe Interagency Task Force 
on Workplace Safety and Health, created by President Carter in 
August lOTT.^'-^^ Much of the impetus for the Task Force came from a 
memo to the President from Charles Schultze (chairman of the Coun- 
cil of Economic Advisers), Stuart Eizenstadt (White House Staff), 
and Bert Lance (then head of OMB) that argued for the use of eco- 
nomic incentives to promote occupational safety. The Task Force — 
which includes representatives from the Council of Economic Advisers, 
OMB, the Small Business Administration, and the departments of 
Commerce, HEW. and Labor— will focus on three issues: (1) '^Ex- 
ploration of incentives that might supplement workplace safety regu- 
lations," where incentives are broadly defined to include education, 
information, tax breaks for employers who improve safety, workers' 
compensation, and "deterrent penalty structures" (e.g., an injury tax) ; 
(2) "Evaluation of the government-wide administration of Federal 
workplace safety and health activities"; (3) "Review of other ways 
to improve the safety and health efforts of all Federal agencies." "2 

It is heartening that the Task Force will examine a broad range of 
options. Its charge to explore the use of incentives is particularly wel- 
come. One can only wish that such a study had been done 7 years ago, 
before OSHA embarked on its present i ourse of action. The author- 
ity of the Task Force, however, is limited to making recommenda- 
tions. Given the course that OSHA has been charting for itself this 
past year, it is unlikely that the Task Force's findings will lead to 
any fundamental redirection of OSHA's efforts, although there may 
be marginal improvements. 

Virginia ^jfxnuTry''v>^''w7l^^^ ^''''''^ ""^ Workplace Safety and Health, Work Plan, Rosslyn 

222 Ibid., p. I-i. "" 



OSHA, not unconscious of its poor public image, has responded ef- 
fectively under Dr. Bingham's leadership to the oft-heard criticism 
that many of its safety standards are nonproductive yet burdensome. 
Increasingly the Agency is applying its resources to the protection of 
workers' health, an area at least as important as workplace safety. 
This development would be welcome if OSHA were to formulate its 
health interventions on a rational basis, considering both the costs and 
benefits of its actions. Unfortunately, OSHA's two most recent major 
health proposals, the new permanent standard for benzene and the 
generic approach to the regulation of carcinogens, suggest that the 
Agency has little interest in analyzing either costs or benefits before 
it intervenes. The result may well be the expenditure of enormous sums 
in return for little or no improvement in workers' health. Recent ex- 
perience reinforces our earlier conclusion : Congress should relegislate 
OSHA's mandate, requiring the Agency to pa.y explicit attention both 
to the health benefits that its interventions generate and to the resource 
costs that they entail. Such a requirement would encourage OSHA to 
rationalize its interventions, directing them to the areas where they 
would be most productive. 


Arthur D. Little, Inc. Technology Assessment and Economic Impact Stndy 
of an OSHA Regulation for Benzene. Draft report prepared for U.S. Department 
of I.abor, OSHA, undated. 

Ashford, N.A. Crisis in the Workplace. MIT Press. Cambridge, Mass.. 1076. 

Ashford, N. A. "An Evaluation of the Occupational Health Mandate of the Oc- 
cupational Safety and Health Act," presented at a U.S. Department of Labor 
Conference on "Evaluating the Effects of the Occupational Safety and Health 
Program," Annapolis, Md., Mar. 18-19, 1975. 

Ashford, N. A. et al. "Economic/Social Impact of Occupational Noise Regula- 
tloins," testimony presented at the OSHA hearings on the economic impact of 
occupational noise exposure, Washington, D.C., Sept. 30, 1976. 

Baumol, W. J. and Oates, W. E. "The Use of Standards and Prices for Pro- 
tection of the Environment," in P. Bohm and A. V. Kneese (ed.). The Economics 
of Environment, London, 1971. 53-65. 

Bolt, Beranek & Newman, Inc. "Economic Impact Analysis of Proposed Noise 
Control Regulation." Report prepared for the U.S. Department of Labor, Occu- 
pational Safety and Health Administration, Apr. 21, 1976. 

Bur('au of National Affairs. The Job Safety and liealtb Act of 1970. Bureau 
of National Affairs, Washington, D.C., 1971. 

Bureau of National Affairs. "OSHA and the Unions : Bargaining on Job 
Safety and Health." Bureau of National Affairs, Washington, D.C., 1973. 

Calabresi. G. Tbe Cost of Accidents: A Legal and Economic Analysis. Yale 
University Press, New Haven, 1970. 

Chelius, J. R. "The Control of Industrial Accidents: Economic Theory and 
Empirical Evidence," Law and Contemporary Problems, 38 (1974). 700-29. 

Chelius, J. R. "Expectations for OSHA's Performance : The Lessons of Theory 
and Empirical Evidence," Purdue University, March 1975. 

Coase, R. H. "The Problem of Social Cost," Journal of Law and Economics, 
3 (1960), 1-44. 

Cornell, N., Noll, R., and Weingast, B. "Safety Regulation," H. Owen and C. E. 
Schultze (eds.). Setting National Priorities, Brookings Institution. 1976. 

DiPietro, A. "An Analysis of the OSHA Inspection Program in Manufacturing 
Industries 1972-73," Draft Technical Analysis Paper, Office of the Assistant 
Secretary for Policy, Evaluation, and Research, U.S. Department of Labor, 
August 1976. 

DiPietro, A. "Data Needs for the Evaluation of OSHA's Net Impact," Office 
of Evaluation, Office of Assistant Secretary for Policy, Evaluation, and Research, 
U.S. Department of Labor, Aug. 19, 1975. 

Dunkelberg, W. C. "OSHA Activity in the Small Business Community," Draft 
Technical Analysis Paper, Office of the Assistant Secretary for Policy, Evaluation, 
and Research, U.S. Department of Labor, April 1976. 

Dunlop, J. T. "The Limits of Legal Compulsion," unpublished paper, Nov. 11, 

Foulkes, F. "Learning to Live with OSHA," Harvard Business Review (No- 
vember-December 1973). 

Hale, A. R. and Hale, M. A Review of the Industrial Accident Research 
Literature. HM's Stationery Office, London, 1972. 

Infante. P.F. et al. "Leukemia Among Workers Exposed to Benzene." National 
Institute for Occupational Safety and Health. Cincinnati, Ohio, April 28, 1977. 

Job Safety and Health, various issues. 

Johnson, F. C. "Workers' Compensation Laws— Changes in 1974," Monthly 
Labor Review, 98 (January 1975), 30-37. 

Kalis. D. C. "T^pdate on the Cost of OSHA (Compliance," Occupational Haz- 
ards, August 1974, 42-44. 

Kasper, D. M. "An Alternative to Workmen's Compensation," Industrial and 
Labor Relations Review, 28 (1975) , 535-48. 

8.3-944—78 17 


Kneese, A. and Schultze, C. Pollution, Prices, and Public Policy. Brookings 
Institution, Washington, D.C., 1975. 

Kunreuther, H. "Limited Knowledge and Insurance Protection," Public Policy, 
24 (1976), 227-62. 

Larson, A. The Law of Workmen's Compensation. 5 vol. (looseleaf). M. Bender, 
Albany, N.Y., 1952-76. 

Mendeloff, J. "An Evaluation of the OSHA Program's Effect on Workplace 
Injury Rates : Evidence from California Through 1974," Report prepared for the 
Office of the Assistant Secretary for Policy, Evaluation, and Research, U.S. De- 
partment of Labor, under contract B-9-M-5-2399, July 1976. 

Miller, R. S. "The Occupational Safety and Health Act of 1970 and the Law 
of Torts," Law and Contemporary Problems, 38 (1974) , 612^0. 

Morey, R. S. "The General Duty Clause of the Occupational Safety and Health 
Act of 1970," Harvard Law Review, 86 (1973) , 98S-1005. 

National Commission on State Workmen's Compensation Laws. Compendium 
on Workmen's Compensation. Washington, D.C., 1972. 

National Commission on State Workmen's Compensation Laws. Report of 
the National Commission on State Workmen's Compensation Laws. Washington, 
D.C., July 1972. 

Oi, W. Y. "On the Economics of Industrial Safety," Law and Contemporary 
Problems, 38 (1974), 669-99. 

Oi. Walter Y. "On Evaluating the Effectiveness of the OSHA Inspection Pro- 
gram," Report submitted to U.S. Department of Labor, Contract No. L-72-86, 
May 15, 1975. 

Page, J. A. and O'Brien, M. Bitter Wages. Grossman, New York, 1973. 

Pfeifer, C. M. et al. (eds.) "An Evaluation of Policy Related Research on 
Effectiveness of Alternative Methods to Reduce Occupational Illness and Acci- 
dents," Report prepared by Behavioral/Safety Center, Columbia, Md., July 1974. 

The President's Report on Occupational Safety and Health. Annual reports 
for 1972, 1973, and 1974. 

Prosser. W. L. Handbook of the Law of Torts. 4th ed.. West Publishing Co., 
St. Paul, Minn., 1971. 

Ruff, L, "The Economic Common Sense of Pollution," Public Interest (spring 
19,70), 69-85. 

Russell, L. "Safety Incentives of Workmen's Compensation Insurance," Jour- 
nal of Human Resources (summer 1974) , 361-75. 

Sands, P. E. "How Effective is Safety Legislation?" Journal of Law and Eco^ 
nomics, 11 (April 1968), 165-74. 

Schultz, C. E. "The Public Use of Private Incentives," Godkin Lectures, Har- 
vard University, Nov. 29-Dec. 7, 1976. 

Settle, R. "Benefits and Costs of the Federal Asbestos Standard," paper pre- 
sented at Department of Labor Conference on Evaluating the Effects of the Occu- 
pational Safety and Health Program, Annapolis, Md., Mar. 18-19, 1975. 

Settle, R. F. and Weisbrod, B. A. "Governmentally Imposed Standards : Some 
Normative and Positive Aspects," unpublished paper, undated. 

Shafai-Sahrai, Y. Determinants of Occupational Injury Experience, Divi- 
sion of Research, Graduate School of Business Administration, Michigan State 
University, East Lansing, 1973. 

Singer, J. W. "Labor Report/New OSHA Head May Signal Change in Agency's 
Approach," National Journal (Dec. 27, 1975 ), 1725-34. 

Singer, J. W. "New OSHA Task Force — Political Payoff or False Alarm?" Na- 
tional Journal (July 10, 1976), 973-75. 

Singer, J. W. "A New OSHA Tried to Put Its Pieces Back Together Again." 
National Journal. July 2, 1977, 1040-49. 

Smith, A. The Wealth of Nations. Modern Library, New York, 1937. 

Smith, R. S. "The Estimated Impact on Injuries of OSHA's Target Industry 
Program," Cornell University, undated. 

Smith, R. S. "Evaluating the Impact of OSHA on Occupational Safety and 
Health," Office of the Assistant Secretary for Policy, Evaluation, and Research, 
U.S. Department of Labor, April 1975. 

Smith, R. S. "The Feasibility of an 'Injury Tax' Approach to Occupational 
Safety," Law and Contemporary Problems, 38 (1974), 730-744. 

Smith, R. S. The Occupational Safety and Health Act. American Enterprise 
Institute, Washington, D.C., 1976. 

Stender, John H. "Enforcing the Occupational Safety and Health Act of 1970 : 
The Federal Government as a Catalyst," Law and Contemporary Problems, 38 
(1974), 641-650. 


Surry, J. Industrial Accident Research: A Human Engineering Appraisal. 
Xabour Safety Council, Ontario Department of Labour, Toronto, 1971. 

Thaler, R. and Rosen, S. "The Value of Saving a Life: Evidence from the 
Labor Market," paper presented at NBER Conference on Income and Wealth, 
Household Production and Consumption, Washington, D.C., Nov. 30, 1973. 

Tversky, A. and Kahneman, D. "Judgment Under Uncertainty: Heuristics and 
Biases," Science, 185 (1974), 1124-1131. 

U.S. Bureau of Labor Statistics, "Major Collective Bargaining Agreements in 
Selected Industries : Safety and Health Provisions Before and After the Occu- 
pational Safety and Health Act of 1970." Report prepared by the Office of Wages 
and Industrial Relations, for the Office of Occupational Health Statistics, Bureau 
of Labor Statistics, Washington, B.C., February 1975. 

U.S. Congress, House of Representatives, Committee on Education and Labor, 
Select Subcommittee on Labor, Occupational Safety and Health Act of 1970 
(Oversight and Proposed Amendments). Hearings (93d Cong. 2d sess.), Wash- 
ington, D.C., 1975. 

U.S. Congress, House of Representatives, Permanent Select Committee on 
Small Business, Subcommittee on Environmental Problems Affecting Small 
Business, The Effects of the Administration of the OSHAct on Small Business. 
Hearings (93d Cong., 2d sess.), Washington, D.C., 1974. 

U.S. Congress, Senate, Committee on Labor and Public Welfare. Legislative 
History of the Occupational Safety and Health Act of 1970, Washington, B.C., 

U.S. Congress, Senate, Committee on Labor and Public Welfare, Subcommittee 
on Labor. Occupational Safety and Health Act Review, 1974. Hearings (93d 
Cong., 2d sess.), Washington, B.C., 1974. 

U.S. Bepartment of Health, Education, and Welfare. "Occupational Exposure 
to Coke Oven Emissions." U.S. BHEW National Institute for Occupational 
Safety and Health, Washington, B.C., 1973. 

U.S. Bepartment of Health. Education, and Welfare. "Occupational Exiwsure 
to Inorganic Arsenic: New Criteria — 1975." U.S. BHEW, National Institute for 
Occupational Safety and Health, 1975. 

U.S. Bepartment of Labor. "Inflationary Impact Assessment : 2-Butanone." 
Occupational Safety and Health Administration, Washington, B.C., May 25, 1976. 

U.S. Bepartment of Labor. "Inflationary Imi)act Assessment : Commercial 
Biving." Occupational Safety and Health Administration, Washington, B.C., 
September 197G. 

U.S. Bepartment of Labor. "Inflationary Impact Assessment : Cyclchexanone." 
Occupational Safety and Health Administration, Washington, B.C., May 25. 1976. 

U.S. Bepartment of Labor. "Inflationary Impact Assessment : Ethyl Butyl Ke- 
tone." Occm)ational Safety and Health Administration, Washington, B.C., May 
25, 1976. 

U.S. Bepartment of Labor. "Inflationary Impact Assessment : Hexone." Occu- 
pational Safety and Health Administration. Washington, B.C., May 25. 1976. 

U.S. Bepartment of Labor. "Inflationary Impact Assessment : Methyl (n-Amyl) 
Ketone." Occupational Safety and Health Administration, Washington, B.C., 
May 25, 1970. 

U.S. Bepartment of Labor. "Inflationary Impact Statement : Coke Oven Emis- 
sions." Occupational Safety and Health Administration, Washington, B.C., 
Feb. 27. 1976. 

U.S. Bepartment of Labor. "Inflationary Impact Statement: Inorganic Ar- 
senic." Occupational Safety and Health Administration, Washington, B.C., 

U.S. Bepartment of Labor. "Occupational Safety and Health Administration's 
Impact on Small Business." Occupational Safety and Health Administration, 
Wasbington, B.C., undated. 

U.S. Interagency Task Force on Workplace Safety and Health. Work Plan. 
Rosslynn, Virginia, January 12, 1978. 

Viscusi, W. K. Employment Hazards: An Investigation of Market Perform- 
ance, Ph. B. dissertation. Harvard University, Cambridge, Mass., 1976. 

Warren. E. W. et al. "Post Hearing Brief of API, NPRA and Individual Com- 
panies." Kirkland, Ellis & Rowe. Washington, B.C., September 22, 1977. 

"What It's Costing Industry to Comply with OSHA," Occupational Hazards, 
fall, 1974. 

Zeckhauser, R. "Procedures for Valuing Lives," PubUc Policy, 23 (1975), 419- 


Individuals Interviewed fob this Report 

Morton Corn, Assistant Secretary of Labor for OSH, Washington, D.C. 

Aldona DiPietro, economist, Office of the Assistant Secretary for Policy, Evalu-i 
ationj and Research, Department of Labor, Washington, D.C. 

Frank Frodyma, Acting Chief, Division of Programing and Systems Analysis, 
OSHA, Washington, D.C. 

Harlan Holdefer, member, Field Performance Evaluation Team, OSHA, Wash- 
ington, D.C. 

Dan Jacoby, Counsel for the Office of Standards, OSHA, Washington, D.C. 

Jack Katalinas, Director, Office of Management Data Systems, OSHA, Wash- 
ington, D.C. 

Anson Keller, Special Assistant for Regulatory Affairs, OSHA, Washington, 

Joe Kirk, Director, Policy Analysis and Integration, OSHA, Washington, D.C. 

Joe Nolan, member, Field Performance Evaluation Team, OSHA, Washing- 
ton, D.C. 

Gilbert Saulter, regional administrator for OSHA, Boston, Mass. 

Grover Wrenn, Chief, Division of Health Standards Development, OSHA, 
Washington, D.C. 

(By Larry E. Ruff) 


This paper deals with a limited but important aspect of environ- 
mental policy, namely, that part which seeks to manage the quality 
of society's air and water resources by controlling "pollution." The 
current and emerging legislation dealing with this part of environ- 
mental policy — the Clean Air Act, the Federal Water Pollution Con- 
trol Act, the Toxic Substances Control Act — all emphasize a regula- 
tory approach, as do most proposals for revision and extension of this 
legislation. In a sense, then, this paper deals with environmental 
recjulation^ even though its principal theme is that environmental 
policy should rely less on regulation and more on other resource man- 
agement tools, particularly those which use economic forces to help al- 
locate valuable social resources. 

The paper identifies an important distinction between environ- 
mental problems and the kind of economic problems with which 
regulation typically deals : environmental regulation is not merely a 
matter of helping an existing market work somew^hat better, but is a 
matter of using administrative mechanisms to allocate a valuable re- 
source in the complete absence of cooperating market forces. Because 
of this, it is not possible to argue logically that environmental prob- 
lems could be "left to the market," as one might choose to leave trans- 
portation or communications problems to the market. But this distinc- 
tion also suggests that relying on pure regulatory methods will be 
unusually difficult, because there is no market which can be corrected 
or guided — regulation must try to perform all the economic functions 
wliich arc usually, and for good reason, left to market forces. Thus, 
wise policy should consider ways in which market or incentive systems 
could be created to help manage these complex and valuable resources. 

The two basic forms environmental markets could take, and some 
of the many possible combinations and variations, are described. Sys- 
tems of discharge price's would impose monetary prices on discharges, 
perhaps varying the price level with time and location when there are 
clear environmental reasons for doing so, but not making distinctions 
among discharges on the basis of individual internal technical or eco- 
nomic factors. Systems of m-arhetahJe discharge Tights would define 
an acceptable total level of discharges in a region, allocate this total 
among dischargers in some way— perhaps by an auction— and then 
let the defined discharge rights be traded in a regulated market under 
certain specified conditions. 

Implementing some combination of these market devices to help in a 
particular environmental management situation would encounter many 
of the same difficulties encountered by regulatory systems — the need 
to estimate discharges, the need to make judgments about how fast and 
to what extent environmental improvement should be pursued, the 
need to decide whether some discharges should be treated as though 
they are more harmful than others. But by reducing the need for pub- 



lie officials to make and defend judgments about the motives, tech- 
nology, and economic situation of each individual pollution source, 
these market systems could substantially ease the administrative bur- 
den and help improve the effectiveness, efficiency and equity of pollu- 
tion control programs. 

The history of environmental policy at the federal level in the U.S. 
is reviewed. This history suggests that efforts to make regulatory 
systems of environmental management successful have made them less 
flexible, more demanding, less able to make cost-effective distinctions, 
and more arbitrary in their definitions and deadlines. It is suggested 
that the 1970 Clean Air Act (CAA) and 1972 Federal Water Pollu- 
tion Control Act (FWPCA) pushed this trend about as far as it is 
possible to go. As these Acts are revised in the light of reality, they 
can only revert toward the traditional pattern of regulatory discre- 
tion and bureaucratic complexity — unless fundamentally different ap- 
proaches are tried. 

The implementation of the 1970 CAA and 1972 FWPCA as they 
have been amended through 1977, and the prognosis for the future if 
no significant changes are made, are discussed. It is shown that the 
elaborate regulatory procedures of both programs have not worked 
as planned. In effect, both programs have become little more than mas- 
sive negotiating operations, in which any polluting act can be defined 
to be legal or illegal, depending on many complex factors, and the 
regulatory agencies and the polluters try to work out a modus vivendi 
on a case-by-case basis. Although some short-term progress is being 
made, the situation does not suggest that sound programs for man- 
aging environmental resources in the long run are being put in place. 

On the basis of experience with these programs to date and other in- 
formation, some lessons with important policy implications can be 
learned. Among these is that, for most pollutants, it is unlikely that 
any curve relating discharges to damages in a real region of the world 
will exhibit any "thresholds" or "jumps" which can be used as an 
objective policy goal ; a "linear assumption" of a constant rate of in- 
crease of damage with pollution is probably as good as any other, 
within the limits of uncertainty and variability. In other words, par- 
ticular quantitative pollution goals and timetables cannot be given 
much sanctity. The control cost curve, on the other hand, typically will 
be steep at the high levels of control being demanded, and the attainable 
degree of control will be uncertain and variable for each source, mak- 
ing quantitative discharge limitations difficult to set. Taken together, 
these generalizations about control cost and environmental damage 
curves suggest that the best policies are those which let technology and 
economics, rather than some preselected environmental goal and dead- 
line, determine the rate and extent of discharge reductions — discharge 
prices are better than marketable discharge rights, and best-technology 
regulations are better than inflexible ambient standards. Although 
there surely are exceptions to this generalization, it is particularly 
valid in the early stages of an environmental program, when little is 
known about the best ultimate solution and the immediate need is to 
begin making progress toward lower discharges in a cost-effective 
manner over time. 

Unfortunately, most of the enacted and proposed modifications in 
the CAA and the FWPC^A are in the wrong direction. Arbitrary am- 


bient indices are being increasingly relied upon in some cases. 'Where 
greater discretion and flexbility is being given to the regulatory agen- 
cies, few innovative implementation devices are being provided to help 
maintain the effectiveness of the programs. Where market-like devices 
have been suggested, such as bridge tolls or parking surcharges for 
transportation controls. Congress has acted explicitly to prevent their 
implementation at the federal level. 

This experience and more general observations suggest that market 
systems have far to go to overcome the political opposition to them. 
Xeither the most avid advocates of polUition control nor the pollute i^ 
tliemselves are anxious to trade administrative processes — which they 
both expect to use to their advantage, the former to punish the evil 
polluters, the latter to obtain protection — for the impersonal working 
of market forces. Nevertheless, in the last section some suggestions are 
•offered for steps which could be taken within the context of current 
programs to implement environmental markets. 


(By Larry E. Eutf) 

The Environment, the Eegulatory Process and the Market 

The policy response to the emergence of environmental problems in 
the U.S. has been straightforward : the traditional processes of regula- 
tion have been applied with a minimum of modification. But there 
are significant differences between environmental problems and the 
economic difficulties which have led to economic regulation in the past. 
These differences and their implications for traditional regulation are 
discussed in general in this section. Then the ways in which market 
forces might be used to help overcome these difficulties are outlined. 

In the U.S., it has traditionally been accepted that market forces of 
competition, supply and demand, profit and loss are effective and 
powerful devices for stimulating and guiding economic activity. ^Hien 
it has been perceived that thcvSe forces were not working properly, or 
were producing unacceptable results, public policy has typically tried 
to make them work better rather than to get along without 
them, and has taken the form of "regulating" the existing markets so 
that they perform more like markets should perform. For environ- 
mental problems, however, this approach may be less appropriate and 
less effective than in other areas. The reasons for this and the policy 
implications are discussed here, and then the general process of envi- 
ronmental regulation is outlined. 

The Nonexistence of Environmental Markets 

Economic regulation is a well-developed process which attempts to 
improve the functioning of existing markets in the economv. Thus, 
price and service regulation has been imposed on natural monopolies 
m an effort to get them to act like competitive firms; price and entry 
regulation has been imposed on industries such as transportation for 
purposes including eliminating destructive competition; licensing 
and minimum-standard regulations have been instituted where it is dif- 
ficult for buyers to become adequatelv informed; tax and subsidy de- 
vices have been used where it was felt necessary to modify market 
forces; and anti-trust laws have attempted to foster competition— all 
m an effort to make existing markets function more like they should. 

hov our purposes, the critical feature of tliose situations m which 
economic regulation has traditionallv been attempted is that a market 
already exists, and continues to plav a central role, so that the regula- 
tor need only worry about things at the margin. The CAB may have a 
legal mandate to promote and regulate commercial aviation ; but it is 
the power of economic forces, more than the managerial genius and 



regulatory power of the CAB, wliich gets aircraft built, keeps supply 
in line with demand, decides what service should be offered, and the 
like. The Consumer Product Safety Commission may be responsible 
for keeping dangerous products off the market ; but it is the fact that 
consumers themselves prefer safety to danger and act in the market 
(and the courts) accordingly, more than the diligence and omniscience 
of the CPSC, which prevents major carnage. Similarly with the FDA, 
the ICC, the FCC, and other regulatory agencies : the positive and dy- 
namic features of the industry, the generally-satisfactory nature of 
transactions in it, come about because market forces push in that di- 
rection ; successful economic regulation tries to provide some stability, 
prevent some mistakes, and take care of some marginal inefficiencies 
and inequities, without interfering too much with the normal workings 
of the market. 

Environmental problems, however, are quite different in form. The 
"market failure" which causes an environmental problem is not be- 
tween buyers and sellers in the market for pollution control equip- 
ment, or between buyers and sellers of the steel or paper which comes 
from the polluting mills ; rather, it is between the "buyers" and "sell- 
ers" of the environmental resources themselves. Clean air, clean water, 
wilderness areas, and the earth's ozone layer are resources which pro- 
vide nature and human society with a whole array of indispensable 
services, in support of recreation, industry, agriculture, health, and the 
very life-sustaining processes of the earth. "When human society puts 
few demands on these resources, the supply is adequate for all. But as 
human numbers, concentrations and activities grow, the demands on 
environmental resources begin to compete and interfere with one 
another: the demand for air to carry away wastes can no longer be 
met without impairing scenic vistas, the health of breathers, or the life- 
protecting functions of the upper atmosphere ; the demand for water 
to carry away heat or wastes becomes such that demands for swim- 
ming, fishing and drinking cannot be met simultaneously. In short, the 
environmental resources become subject to more competing and im- 
portant demands than they can meet. 

The "normal" process by which such competing and conflicting de- 
mands on a resource are mediated is supply and demand in a market : 
those who own or control the resource let others use it only on terms 
acceptable to the owners, which usually means the users pay for the 
privilege; and the owner, in seeking to maximize the return he gets, 
sells the resource or its services to those offering to pay the most for it, 
and then only if he himself has no more valuable use for it. The result 
is that users of the resource are continually seeking ways to conserv^e 
it, and only those uses for which the resource is most valuable — as de- 
termined by the tastes and income distribution of the society — suc- 
ceed in getting access to the resource. Even such natural resources as 
plants, animals, land and minerals have undergone the historical trans- 
formation from free and abundant gifts of nature to scarce and valu- 
ble resources, and have become accepted parts of the market economy. 

Public policy could and should seek to create some market-type in- 
stitutions for managing environmental resources but (at least in the 
U.S.) has not yet done so. Thus, environmental regulation must cope 
with a situation fimdamentally different from that in which most 


economic regulation takes place: the environmental regulator is not 
trying to correct existing market forces which are basically supportive 
of his goals, because there are virtually no such forces. Buyers of en- 
vironmental quality are not negotiating with sellers who can refuse to 
sell if their terms are not met. There is no automatic market for the 
new deidce or idea which will conserve on the use of environmental re- 
sources. There are not even any prices about which buyers and sellers 
and regulators can argue,' and which can be controlled or adjusted or 
stabilized. The environmental regulator is tr\^ing to manage an ex- 
tremely valuable and critical set of resources with virtually no help 
from market forces. 

There are a number of important implications of this difference be- 
tween environmental regulation and the more-traditional economic 
regulation. The first is that it is not possible to argue logically that en- 
vironmental problems could be "left to the market,*' in the same way 
that, say, transportation or consumer product safetj^ or cable television 
could in principle be "left to the market.-' In these other areas, the issue 
is whether, on balance, regulation improves the functioning of the ex- 
isting market. But there are no markets for environmental resources 
which would automatically deal with the problems if regulatory pro- 
grams were dismantled and nothing put in their place. Of course, it is 
still necessary to think carefully about any pai-ticular environmental 
problem, to be sure it is so serious that it is worth paying the manage- 
ment costs necessary to try to deal with it. And there are policies be- 
sides "regulation" which should be considered. But. for most of the 
environmental problems discussed here, it is not really possible to do 
nothing and simply "leave it to the market." 

A second implication of the difference noted between environmental 
and economic regulation is that, in formulating environmental policy, 
one may need to make much more basic changes in, and be more in- 
novative about, the relationship between government and the econo- 
my. Where the private, market -based economy functions reasonably 
well, it is probably wise to leave it more-or-less alone, doing a mini- 
mum of constraining and regulating, and then generally from out- 
side the market and its actors; i.e., government can accomplish its 
ends by setting out the roles and limits within which the market 
economy operates. However, when the market economy is structurally 
incapable of dealing with an iniportant problem it may be necessary 
for government to get inside the private economy to correct ih^ struc- 
tural flaw, even if this requires some real policy innovation. To the 
extent that environmental problems are caused by the absence of cer- 
tain market forces, it is suggested that policy might take the form of 
simulating these market forces ; it should not be assumed that the best 
solution is just to beef up the traditional tools of economic regulation 
and apply them in a fundamentally different situation. 

A third implication is that if policy does not take the form of pro- 
viding some supporting market forces, the job of the environmental 
regulator is much more difficult than the job of most economic regu- 
lators. The environmental regulators do not get much help from the 
market in developing technology, sorting out priorities, providing 
information about alternatives, stimulating competition, or reducing 
frivolous demands on them or on the environment itself. They quickly 
get drawn into detailed technical and economic issues across a broad 


spectrum of the society. Because so much money hangs on each detail, 
and because delay is worth so much, litigation is extensive and poli- 
tical pressure is intense. Some of this is inherent in the complexity 
of environmental issues, and some of it could be handled by providing 
more staff, budget and independence to the regulators. But a large 
part of it is due to the basic fact that it is just very difficult to use 
the processes of regulation and bureaucratic administration to handle 
the kind of economic problems which are generally, and with good 
reason, left to market forces. 

Regulating Environmental Resources 

The regulatory process of interest here is that which uses the police 
powers of government to compel people to do substantive things which 
they would not otherwise do.^ This process generally works through 
an agency charged with translating general legislative objectives into 
specific program goals, translating these goals into regulations appli- 
cable to identifiable persons, and then enforcing (or recommending 
enforcement of) the regulations. For simple problems, the objectives 
can be readily understood and accepted, the goals can be definite and 
attainable, and the regulations can be simple and not particularly 
onerous; in such cases, those subject to the regulations will have little 
incentive to challenge them and little chance of doing so successfully, 
and regulation works smoothly. In more complex situations, however, 
the objectives may be uncertain and changing, the goals poorly justi- 
fied and difficult to accomplish, and the regulations complex and costly ; 
in these situations, the large body of law which has been developed 
over the years to protect individuals from arbitrary or capricious 
actions of government officials can be used to advantage to delay, 
weaken and generally disrupt the regulatory process. Unfortunately, 
many serious environmental problems are in this latter category. 

The Goals of Enmronmental Regulations 

For environmental resources, the definition of appropriate regula- 
tory goals is particularly difficult. Knowledge of the basic natural 
processes involved is extremely poor: it is not even known for sure 
how to define or measure "pollution" in a way which is clearly rele- 
vant to human concerns, and choosing the "right" level of some chosen 
measure of environmental quality requires difficult trade-offs involv- 
ing personal and social values. If the regulator has been given only a 
general goal, such as "protect the public health and welfare," he will 
be able to provide a clear and defensible case for only the simplest 
of actions, such as removing a "clear and present danger." or requir- 
ing some clearly beneficial action which costs very little. For all those 
other actions necessary to manage environmental resources, where 
there are real costs and ill-defined or uncertain benefits, a vague "pub- 
lic-interest" goal is not much help. 

As this has become evident over the years, the framers of environ- 
mental policy have progressively narrowed and tightened the stated 
goals of environmental regulation, leaving less discretion to the regu- 
latory officials and making it easier for them to demonstrate the rela- 

1 That Is, we are not discussing here procedural or definitional regulations, such as those 
Issued by the Internal Revenue Service. Any environmental program will have regulations, 
If only to require monitoring, release of data, or payment of a tax. For our purposes, a 
"regulatory program" goes beyond this and tells a discharger what he must do about his 


tionship between their regulations and the legislated purposes. This 
trend has revealed the other horn of the regulatory dilemma: the 
"right" goals of environmental policy are, in fact, uncertain, com- 
plex, shifting and evolving, and making them simple and rigid in 
the laws makes it more difficult to accommodate the complexities and 
uncertainties of reality. Tlie fact that, under the U.S. system of gov- 
ernment, legislatures can be more arbitrary and capricious than regu- 
latory officials may make it easier for legislatures to set goals which 
stand up in court, but does not make it any more valid to use a simple, 
rigid goal to represent society's environmental interests. And, as re- 
cent history has known, even legislatively-set goals have a way of 
coming undone Avhen the courts and the body politic begin to doubt 
their wisdom or feasibility. 

Complexity, variability and risk-taking are inherent in environ- 
mental policy, and rough judgments will have to be made in any effec- 
tive environmental program. Furthermore, basic social values are in- 
volved in decisions concerning the environment. As a result, political 
judgments and not just technocratic calcuhition are required in de- 
termining how fast and how far environmental improvement shoukl 
proceed. But logic and history suggest that the regulatory process, in 
order to be effe<;tive without granting unreasonable discretion to the 
regulatory officials, requires specific, simple, preferably quantitative 
statements of goals, together with rigid deadlines; e.g., pollutant con- 
centrations must not exceed X by 1975. These goals and deadlines are 
due more to the exigencies of the regulatory process than to the broader 
needs of society and, as a result, can be maintained in the long run only 
by imposing unnecessary costs on society. Thinking of environmental 
problems in terms of other than pure regulation might allow programs 
to be developed which can be effective without ignoring so much of the 
complexity, variability and uncertainty of the world. 

The Enforceable Rules 

The specific goals of a regulatory program, even when stated in 
precise, quantitative terms and tightly constrained by the legislation, 
are seldom directly enforceable on persons. The regulatory agency 
must translate the goals into a set of rules and regulations specifying 
what each person must do and when in order to be in compliance with 
the law ; e.g., discharges must not exceed X pounds per hour by 1977. 
For an environmental regulator with no help from market forces, this 
process is equivalent to trying to force by regulations the kind of econ- 
omizing on environmental resources which would result if there were 
a market in these resources. The regulator must decide how much of 
the valuable resources should be used by each person who does or might 
want to use them, and then must write a set of regulations which will 
restrict individual usage accordingly. Thus, environmental regulation 
is more akin to distributing public largess than it is to setting the 
ground rules within which the market will operate — typically with 
the added difficulty that the largesse has already been given away, 
and the regulator is trying to compel some persons to return some of it. 

Because the resources are valuable and are available free of charge 
to anyone who is able to convince the reguLator (or the hearing board 
or the court or the legislature) that he really "needs" them, there is 
great pressure put on the sj^stem. If the regulatory objective includes 


a desire to accomplish a reasonably efficient allocation of the environ- 
mental resource, the regulations must make all manner of distinctions, 
on the basis of subtle differences in technology, economics, location, 
od infinitum^ and translate these into justifiable differences in regula- 
tory requirements. When the stakes are high and the issues complex, 
each of these many decisions will be challenged. Even if the challenges 
do not succeed in proving the regulatory agency "wrong," they can 
succeed in delaying and tying up the process. The agency can function 
only by keeping its regulations sufficiently lax that the regulated see no 
gain in challenging them. 

As this problem has become apparent to the framers of environ- 
mental policy, they have taken the only steps possible within a purely 
regulatory framework: the regulators have been directed to make 
fewer distinctions, even at the cost of great inefficiency in accomplish- 
ing the goals; the criteria on which regulations can Jdc justified have 
been made simpler; and legal standing has been granted to those on 
the other side of the (non-existent) environmental market, i.e., the 
"public interest" groups, so that the regulators will not be able to es- 
cape their dilemma by being lax. As with the establishment of en- 
vironmental goals, however, forcing the regulations to be simple when 
reality is complex merely exchanges one set of problems for another. 

In addition to the general problem of trying to decide what each of 
many pollution sources should do to reduce its pollution, there are two 
specific types of problems with which environmental regulation has 
had difficulty at the level of translating legislatively-defined goals into 
enforceable regulations : problems which require the development and 
application of "new" technology ; and problems involving the actions 
of many individuals, e.g., automobile drivers. Even when the legisla- 
tive language suggests that no excuse will be accepted for not accom- 
plishing the goals, it has proven legally and politically difficult to im- 
pose regulations which can be challenged on the grounds that "tech- 
nology is unavailable," or which require substantial changes in the 
behavior of the public. In both cases, the basic problem is that there 
has not been and still is not any market working to control access to 
the environment — polluting is free to those who can get away with it. 
As long as this is true, polluters have every incentive to show no 
imagination in developing solutions to their problem.s, even where only 
minor reengineering and adaptation is required ; and each individual 
member of a group, even while recognizing that group behavior should 
change, will have no incentive to change his individual behavior. Try- 
ing to write and enforce regulations which compel ingenuity and 
initiative, or which can control each of many diverse individuals, is 
no easy task. These problems are dealt with more fully below. They are 
mentioned here to emphasize the point that thinking of environmental 
policy purely in terms of regulation may make it unnecessarily difficult 
to deal with important aspects of the problem. 

Enforcement of the Rules 

Once the goals of a regulatory program have been translated into 

specific rules and regulations applicable to identifiable legal entities, 

these rules are enforced in the courts : an alleged violator is charged by 

the regulatory agency with a violation, is taken to court, and tried; 


the courts mav order the alleged violator to cease his violation imme- 
diatelv or on some schedule, may impose a fine or ]ail sentence, or may 
find the recrulation to be procedurallv improper or substantively un- 
rea<=onablerOver time, the regulations are refined, everyone learns 
Tvhat the courts will allow regulators and regulated to get awav with 
and the program settles down and runs smoothly, under an established 

bodv of law. . -^ .• i 

How well all this really works m a specific regulatory situation de- 
pends on many things, principally how complex and demanding the 
recrulations are. and whether the regulations can be allowed to "settle 
down'- over time or must be continually revised and strengthened. 
When the regulations are simple, violations are easy to find and prove, 
and compliance or non-compliance is a matter of will more than of 
feasibility, then enforcement can proceed as rerrulatory theory says it 
should. But when the regulations are so complex that it is not clear 
what any particular person must do to comply, when noncompliance is 
not easy^to demonstrate, or when there can be any number of technical 
,;nd economic excuses for non-compliance, then enforcement is a diffi- 
cult and time-consuming process. If, in addition, the regulations them- 
selves are changing continually in pursuit of an evolving goal or in 
light of unfolding information, enforcement— in the sense of using 
the law to compel compliance with a set of well-defined and court- 
tested regulations — becomes impossible. 

Many regulatory programs face the problem of complex and shift- 
ing regulations, biit none more seriously than environmental regula- 
tion. Because of the complexity of the situation, the "enforceable" en- 
vironmental rules are often little more than statements to the efi'ect 
that too luich pollution is unlawful, with ''too much" being a matter 
of scientific, technical and economic judgment in each case. The result 
is that the enforcement process becomes a bargaining exercise, with the 
regulatory agency and the individual polluters trying to reach some 
sort of negotiated outcome which the airency can call compliance and 
the polluter can live with. So the polluter makes promises and the 
agency announces victory. When the promises are not kept, renegotia- 
tion takes place and the process repeats itself. 

In all this, a regulatory agency correctly perceives that forcinir an 
issue to litigation is a costly and hazardous move, to be taken only as a 
last resort. As long as a regulation is on the books, the agency can 
negotiate vrith the re'^nilated in a cooperative spirit, trying to work out 
a program which will make reasonable progress toward the agency's 
goals. But once the agency takes an alleged A'iolator to court, coopera- 
tion is replaced with open conflict, and the issue grows beyond finding 
a reasonable solution, as the regulated parties attack the regulation, 
the agency, the goals of the program, and anything else which is liable 
to attract the court's attention. Because the courts tend themselves to 
look for reasonable compromises in such cases, the likely outcome is a 
court-negotiated settlement, along the lines the agency itself would 
have liked to negotiate. Even worse, the courts may throw out the 
regulations, castigate the agency for sloppy work, and set back the 
program substantially. Even if the letter of the regulations is upheld, 
the result may be a political backlash against the program. Thus, the 
regulatory agenc}^ tries to avoid litigation, preferring to work out 

83-944—78 18 


reasonable compromises and to make reasonable progress toward the 

This -theme of environmental regulation as a bargaining process has 
been well-developed by others,^ who tend to view it as not necessarily 
bad, probably beneficial if done well, and certainly inevitable. It is 
generally conceded, however, that it does not produce dramatic prog- 
ress toward environmental goals, because those being regulated will 
find resistance more profitable than cooperation if the regulators push 
too hard. As concern over environmental problems has grown, it has 
become increasingly clear that the effectiveness of environmental 
policy must be improved, and elaborate regulatory schemes have been 
designed to try to put regulation on a more logical, enforceable basis, 
so that it does not have to rely on case-by-case negotiation. Essentially, 
this can be done only by simplifying the goals, using uniform rules, 
and making it clear to the courts that other considerations are not to 
interfere with accomplishment of environmental goals, i.e., by putting 
the needs of the regulatory process above the broader needs of society. 


If markets — even highly imperfect ones — did exist for environ- 
mental resources, the difficulties of environmental management could 
be substantially reduced. The environmental authorities could focus 
their attention and resources on understanding and monitoring en- 
vironmental systems and their interaction with human affairs, defin- 
ing and enforcing the constraints within which the environmental 
markets would function, taking administrative steps or recommend- 
ing legislative action where necessary to help the markets work bet- 
ter, but basically leaving it to the market forces to determine the 
technical and economic details of the when, where, who and how of 
discharge reductions. 

If environmental markets did exist, they would work by requiring 
those who "use up" environmental resources, i.e., who degrade them 
so there is less available for other uses, to pay for the privilege. These 
niarkets^ do not exist because of the difficulty of defining "property 
rights" in environmental resources and defending them against those 
who would encroach on them. Unlike land or animals, which have 
undergone the historic transformation from free goods to economic 
commodities, environmental resources cannot usually be appropri- 
ated and defended— and hence "owned"— by individuals. But if pub- 
lic policy were to seek to define and defend' such property rights, and 
to provide the institutional and legal framework within which these 
rights could be traded, then some of the advantages of markets could 
be realized in managing valuable environmental resources. 

^ Environmental markets established by public policy could take 
either of two basic forms, with any number of variations and com- 
binations possible to accommodate the details of a specific situation. A 
system of discharge prices could be used, in which the processes of 
government would decide the prices which must be paid by any source 

Nn ^r r^^iSwT ?^^<5«,^' ^^' /'Pollution Control as a Bargaining Process." Publication 
^2<i^' ^^^^""^^ University Water Resources Center, Ithaca. New York. 1966; Irwin and 
T,tlW;nf? wi'^i,'?^''? ""^ ?f?.'''■^^n^°/ ^}^^^ ^=^*^^ Pollution Coutrols," Environmental Law 
SnfnSi!: jy^'S'*"-^^??,' P-^- ^?,^^ ' "Economic Law Enforcement. Vol. II." Connecticut 
Enforcement Project," Connecticut Department of Environmental Protection 1975 


Tvhich discharges pollutants of certain types at various times a,nd 
places leaving it to the resulting economic disincentive to determine 
hoAv fast andliow far the discharges will be reduced. Or a system of 
niarketaUe discharge rights could be implemented, m which the proc- 
esses of government would decide how much of which discharges will 
be allowed in an area, would define a set of certificates or rights or 
permits which limit total discharges to the chosen level, and would 
then allow these rights to be traded in a (regulated) market. Often, 
the best system will be some combination, in which a set of marketable 
discharge rights would be issued, reflecting the desired total level, and 
then a liigh price imposed on all discharges not covered by the rights, 
to accommodate uncertainty and variability. 

It is not possible here to discuss in detail the advantages of these 
market devices relative to each other or to the more traditional reg- 
ulatory alternatives. But as background for the later discussion of 
U.S. experience with environmental regulation, it is helpful to outline 
briefly the advantages of market forces in general, to discuss the basic 
forms in which they might be applied to environmental problems, and 
to compare the likely difficulties of implementation with those of 
traditional regulatory systems. 
The Advantages of Erwironmental Markets 

The ai-gument for using market-type devices for managing environ- 
mental lesources is essentially the same as the argument for using 
markets to allocate any valuable resource, from petroleum to nails: 
markets are effective in matching demand to supply, efficient in the 
way they allocate resources, and — given the income distribution of 
the society — equitable in their imposition of costs and benefits. 

Effectiveness in Reducing the Demand for Environmental Re- 

The demand for the waste disposal services of the environment is 
influenced by the same factors which influence how much of anything 
people want to use — the individual cost and inconvenience of getting 
by with less, compared to the individual cost of using more. If the 
price people pay individually for something is too low, in the aggre- 
gate more will be demanded than is available, and some non-market 
device — queues, rationing or regulation — will decide which demands 
go unsatisfied. A high enough price would reduce demand for de- 
structive uses of the environment to any desired degree without the 
need for costly and uncertain rationing mechanisms. Although it is 
hardly conceivable that all decisions about environmental resources 
could or should be made in markets, it would seem to be wise policy 
to at least let market forces handle the gross aspects of reducing de- 
mand for discharges, leaving to regiilation the fine-tuning on the mar- 
gin, as is the case with most economic regulation. 

The most obvious way economic forces can improve the effective- 
ness of environmental management is in making enforcement of dis- 
chai'ge regulations easier. T^'Tien discharges are costly to a dis- 
charger — either because a price must be paid or because a right with a 
niarket value must be held — there is less incentive to argue for delay 
in accoinplishing some specified discharge reduction, less to be gained 
by having equipment break down "unavoidably," less reason to try to 
argue that technology is unavailable. Thus, even if specific discharge 


regulations are maintained, market-like policy devices can improve 
their enforceability. And if the economic pressure is strong enough, 
it will become possible to relax regulatory restraints, maintaining dis- 
charge standards as back-up or emergency powers. 

Even more important than aiding enforcement efforts is the ability 
of market forces to induce environmentally conserving action in the 
long run. By its nature, regulation is limited to forcing relatively sim- 
ple, short-term, easily-identifiable actions. Market forces, by providing 
incentives throughout the economic system, can encourage the kind of 
fundamental and subtle changes in technology, organization and be- 
havior which are necessary if environmental resources are to be effec- 
tively integrated into the ordinary processes of society. It remains trr.e 
that economic self-interest is a more powerful motivator of energy and 
imagination than is exhortation or the force of law. 

Effhciency in A llocating the Supply of Environmental Resources 
The fact that the users of a good or service pay a high price for it is 
good evidence that it is really "needed" by those users who buy it, and 
offers sorne assurance that serious efforts will be made to find new ways 
to economize further on it. The market will allocate the supply 
the competing uses in just about the same way an efficiency-seeking 
regulatory agency would if it had the knowledge and the authority to 
do so, and if it could tolerate the ineffectiveness and inequities whioh 
would result from allowing variability and discrimination in the regu- 
latory process. 

For environmental resources, this efficiency property of markets is 
particularly important in dynamic terms. In a static world, a regu'a- 
tory agency conceivably could gather all the cost information from the 
sources, put it into a computer, and calculate the least-cost way to ac- 
complish some specific environmental objective ; and it might even be 
possible for the regulators eventually to work out payments among 
dischargers so that those who do not undertake control (because con- 
trol is expensive for them) , compensate those who do (because control 
is cheap for them) . 

But in a dynamic world, where technology is continually changing, 
where investment decisions are continually being made and rem.are in 
licfht of unfolding facts, where even the managers of large corporations 
do not try to centralize all information and investment decisions, there 
is no way to technocratically calculate the "least cost" way to accom.- 
plish environmental objectives over time. If market-like policies are 
not used to internalize society's environm.ental objectives into the nor- 
mal processes of the economy, then the environmental regulators must 
get deeply involved in the investment, product, location and technology 
decisions of industry to try to accomplish these environmental objec- 
tives with reasonable ejEciency over time. 

Equity in Allocating Costs and Benefits 

A program which limits discharges of pollutants is, in effect, a pro- 
gram which allocates "rights to pollute" to certain individuals and 
organizations. These rights are valuable, in the sense that they are 
wortli a lot of money to those who get them. Therefore, they should be 
allocated "equitably," in some sense. 

When the discharge rights are given away free of charge and cannot 
be resold — i.e., when traditional regulatory processes are used — equity 


in their allocation is difficult to define, let alone to achieve. Equal allo- 
cations produce unequal benefits, it is not clear what an "equal" allo- 
cation means, and if demands for "equity"' are taken seriously the effec- 
tiveness and efficiency of the decision process can be seriously impaired. 

^Vhen markets are used to allocate the valuable "rights to pollute," 
there is a real sense in which everybody is treated equitably : every- 
body has access to the market on equal terms and pays the same price 
for inflicting approximately the same damage on society. Those 
who use more of the resource pay more for the privilege, those who 
spend money to reduce their use of the resource pay less. The eco- 
nomic value of environmental resources can be captured by the larger 
society which presumably owns them, and need not be given away 
to particular individuals simply because they have good technical 
and economic reasons for using the resources. This economic value 
can be used for general social ])urposes, to compensate those who must 
suffer whatever pollution remains, or to compensate those who are 
felt to have gained, through long and beneficial use, some property 
riglits in the environment. But the felt need to distribute this economic 
value in some particular way need not interfere with the effectiveness 
or efHciency of the programs. 

This concept of economic equity is not the same as broader concepts 
of social justice. But if decisions about who should be allowed to dis- 
rharge how much of what are strongly influenced by notions of social 
( quity or justice, then the decision processes become so complex and 
cumbersome that the effectiveness and efficiency of the policy can be 
seriously jeopardized. And it is not usually the poor, the weak or the 
worthy who learn how to use complex l)ureaucratic and political pro- 
cesses for their own purpose. Therefore, environmental policy should 
■ o allowed to worry "only" about being efficient, elfective and more-or- 

ss equitable in its treatment of people, leaving to other social policies 
; lie broader problems of social justice. 

71ie Basic Forms of Environmental Markets 

There are, of course, many problems involved in the use of environ- 
mental markets. But before discussing these, it is necessary to describe 
in more detail the two basic forms environmental markets might take: 
systems based on discharge prices (or effiuent charges or emissions 
taxes) ; and systems based on marketable disclwrge rights (or saleable 
affluent permits or transferrable emission certificates). 

Marketable Discharge Rights Systems 
Systems based on marketable discharge rights are similar to regu- 
latory systems which attempt to accomplish an "acceptable" total level 
of discharges in a region. Both systems require the environmental 
authority to define the acceptable total discharges for each pollutant, 
to allocate these totals among the individual sources, and to en- 
force compliance with the resulting individual discharge limita- 
tions. The difference is in the way the allowable total discharge 
levels are allocated among the individual sources: the regulatory 
system uses administrative means to give away valuable discharge 
rights which are not transferable, forcing the environmental authority 
to make and remake complex technical and economic judgments about 
-each individual source; the market system defines marketable dis- 


charge rights (MDRs) equal in total to the total allowable discharge 
level, and then lets a market allocate these among the sources. 

One of the strengths of a system based on MDRs is that it maintains 
the control and predictability which are said to be advantages of a 
regulatory system — a public agency decides by how much total pollu- 
tion will be reduced, on what schedule, and uses the force of law to see 
that this total and schedule are met. If it is known that total discharges 
should be reduced from their current level of 100 to a level of 20 
within five years, MDRs can be issued which, in total, allow discharges 
of 100 this year, and 20 in five years, with whatever interim levels are 
thought to be reasonable. If allowable discharges should vary by sea- 
son or weather conditions, the MDRs can specify this. If it is known 
that discharges are more harmful in one location than in another, the 
MDRs can spell out the ratios at which they can be transferred f^'Oin 
one location to another, and transfers can always be made subject to 
approval by the environmental authority. The MDRs can specify al- 
lowable discharges over a year, a day or an hour, or all three and more. 
As discussed elsewhere in this chapter, it is unlikely that such precise 
control is desirable or feasible for most environmental resources, 
whether market or regulatory devices are used. But if such control and 
predictability are desired, and if a regulatory system can be developed 
to deliver them, then a system of MDRs can be incorporated to deal 
with many of the technical and economic details automatically. 

Because a system based on MDRs provides control and predicta- 
bility, it also requires the knowledge and institutional capability to- 
decide rather precisely what the environmental goals and sche lules 
should be, and the ability to enforce specific quantitative discharge 
limitations on individual sources. These difficulties would not be as 
great as they would be under a pure regulatory system, because indi- 
vidual discharo-Prs would have the flexibilitv of buving more MDRs in 
the market if the technology is "unavailable" or if temporary condi- 
tions produce abnormally high discharges. But if the market price 
of the [NrORs becomes too high or their terms are too restrictive, then 
enforcement of them will encounter many of the difficulties typical of 
resrulatory programs — enforcement actions against individual sources 
will be resisted on the grounds that violations were trivial or unavoid- 
able, that the level of available MDRs is unreasonable, etc. Thus, 
MDRs in their pure form are perhaps best suited to situations in 
which the environmental goals and timetables are well-defired, the 
cost of meeting them is known to be acceptable, future develop- 
ments (e.g., an oil crisis) are unlikely to require reconsideration of 
precise goals, and there is a relativelv sophisticated institutional 
structure for gathering information and making the decisions. This 
structure would include regional pollution control authorities. 

Discharge Price Systems 
Systems based on discharge prices are somewhat analogous to regula- 
torv svstems which seek to rednce discharges to a degree and on a schod- 
Tile determined primarily by technology and economics rather than to- 
accomr)lish preset environmental goals and timetables. Once a sched- 
ule of discharge prices is determined Tmore on this presently), dis- 
chargers are free to respond as iih^Y choose, so long as they pav the 
prices. The prices will vary bv nollutant. perhaps with time and locii- 
tion of discharges, but not with the size, product or technology of the 


source. Thus, the environmental agency need concern itself only with 
(environmental impacts, estimating discharges and collecting revenues, 
leaving the individual dischargers to decide how much, how fast, and 
with what technology to reduce their payments by reducing their 

Compared to a system of MDRs, a system of discharge prices allows 
for more variability and uncertainty in total discharge levels and can 
be implemented easily. The environmental authority need not decide 
beforehand how far and how fast the clean-up must proceed, natural 
variability does not produce excusable "violations," and unforeseen 
developments which make it unwise to try to accomplish specific 
quantitative goals need not cause the system to be altered. A simple 
system of discharge prices could be implemented with little more than 
a political judgment to the effect that it is reasonable for all dis- 
cliargers of harmful substance X to pay a price P for the privilege 
of dumping the stuff on the public ; then discharges can be estimated 
with discharge factors, equipment performance ratings, the content 
of raw materials, etc., subject to auditing review, and the revenues 
collected. Thus, discharge pieces are particularly suited to situations 
in which it is known that reductions in discharges are called for. but 
the precise rate and location of the discharge reductions is less critical. 
And compared to a regulatory system which tries to accomplish such 
a goal by imposing "best technology everywhere" standards, the dis- 
charge price requires little direct regulatory involvement in the deci- 
sions and processes of society. 

Hyhrid Systems 
The discharge price and MDR systems can be combined in various 
ways to obtain the advantages of each. A control program for a 
broadly-distributed pollutant can be initiated with a low discharge 
price, which is scheduled to increase over time to a level high enough 
that control will become economical for most sources ; this would begin 
providiuiz information on sources and monitoring methods and an- 
nounoo to existinir and potential '^isr-harofers that they should begin 
immediately looking for ways to reduce their discharges. For some pol- 
lutants, a national or regional discharge price system which encourages 
the application of "host" technology everywhere may be all that is re- 
quired to bring the problem under control. For other pollutants, am- 
bient quality standards or regional limitations on total discharges 
may eventually be promulgated, and sophisticated regional environ- 
mental authorities developed, allowing a system of marketable dis- 
charge rights to be established. These MDRs can be aur^tioned off or 
othcT'wise allocated and then "enforced" bv a high price imposed on 
all discharges in excess of the specified levels, reducing th^ disputes 
over exactly how many MDRs should be issued, whethor excess dis- 
cliarges occurred and are serious enouirh to justifv enforcement ac- 
tions, etc. TVith imagination, hybrid svstems of discharge prices and 
marketable discharge rights could be developed to help in almost any 
environmental management program. 

Some Siibsfantive DliJtnilties in Iw/plernevtation 

The manairoment of environmental resources is a difficult problem 
no matter what policy instruments are used. Knowledge of the natural 
and physical processes involved is poor and always will be. Uncer- 


tainty, variability and evolutionary change are inherent. Thus, it is 
impossible to develop a policy which adheres very closely to any sim- 
ple engineering, ecological, economic or legal model of "rational" en- 
vironmental management. Choices among real policy options must be 
made not on the basis of how closel}' each polic}^ matches its theoretical 
model, but rather on the basis of how well they compare to one another 
in accomplishing society's goals effectively, efficiently, and equitably. 
Suggestions that market devices be used to help manage environ- 
mental resources are too often put to the wrong test in choosing among 
policies. Perhaps because economists have gone to such great lengths 
to demonstrate the logical beauty of market concepts applied to en- 
vironmental resources, the policy suggestions are usually dismissed by 
pointing out the impossibility of implementing them the way economic 
theory says they should be. But the important question is whether, in 
practice, they would have net advantages if used to replace or to com- 
plement the real alternatives based on regulatory processes. Here, 
three of the most difficult problems which would be encountered in 
implementing environmental markets are discussed and compared to 
ihe similar problems of the regulatory alternatives. These problems 
are : estimating discharges ; varying the policy parameters Avith time 
and location; and selecting the level of the price or the number of 

Estimating Discharges 

Any logical system for controlling discharges must have some 
mechanism for estimating physical discharges of pollutants from each 
source, and physical estimates play a central role in present policies. 
Discharge standards are stated in physical terms, and plans for 
achieving ambient goals must use estimates of emissions from each 
source. In many cases these physical estimates come from engineering 
estimates of what a specific teclmology should be able to accomplish, 
the composition of inputs or volume of output, and judgmental guess- 
es rather than from direct measurements of discharges. But estimates 
must be m.ade somehow. 

For monitoring purposes, marketable discharge rights (MDRs) are 
quantitative limitations on individual discharge levels, no different 
from quantitative discharge standards. The uncertainty and impre- 
cision of monitoring device can make it quite difficult to enforce such 
quantitative limitations, because small differences in discharge esti- 
mates may be the only difference between a model citizen and a law- 
breaker. Regulatory programs typically try to convert quantitative 
discharge standards into action plans or compliance schedules, so that 
legal action can be taken for failure to meet a schedule without the 
need for estimating discharges directly. But these plans and schedules 
are, in any rational system, based on estimates of the discharge re- 
ductions they would accomplish, and these estimates could as well be 
used to define and monitor compliance with a set of marketable dis- 
charge rights. As difficult as it is to estimate discharges well enough 
to enforce quantitative discharge limitations, MDRs have no signifi- 
cant disadvantages in this regard relative to administrative standards.' 

3 Of course, If a re^ilatory system can enforce actions, such as use of a narticular I 
technology, without estimatlne the discharge reduction they would produce, this state- 
ment Is not necessarily valid. It Is for the reader to decide whether it Is an advantage or a 
disadvantage of a MDR system that It would require some attention to what physical 
discharges are and how much they would be reduced by various actions. 


In a sj^stem of discharge prices, the estimation problem would be 
somewhat different and probably easier. When a discharge is some- 
thing to be paid for, rather than something subject to regulation, both 
the environmental agency and the dischargers view the estimation 
problem differentlj'. Small differences in estimates cause small dif- 
ferences in payments due, not large differences in fines or expendi- 
tures, reducing the importance of precision and the opportunity for 
serious conflict over small differences — compromise is possible on both 
sides. The environmental agency can begin by using high discharge 
estimates based on industry-wide figures, and leave it to each individ- 
ual source to demonstrate that it is discharging less; this would be a 
complete reversal of the burden-of-proof situation which exists when 
the agency is trying to prove violation of a discharge limitation, 
whethor set administratively or purchased in a market. In fact, nom- 
inal discharge prices, even if too low to be an effective incentive, would 
probably be worthwhile as a means of improving discharge estimates 
under a regulatory system. 

Varying Policy hy Time and Location 

Suggestions for the use of markets in environmental management 
are often rejected because the market proposals do not fit in well with 
the goals and timetables of the current regulatory system, especially 
with respect to the location and timing of discharges. For example, 
the regulatory projrram requires discharges to be reduced by a certain 
date, or to be relatively constant over time without surges, and a 
straight discharge price cannot guarantee this result. Or a marketable 
discharge rights system might allow discharges to become geographi- 
cally concentrated, causing local violation of an ambient quality stand- 
ard, while a regulator}^ system could prevent this. Or a discharge 
price system might allow some regions of the countrv to gain a com- 
petitive advantage over others by charging lower prices. 

These objections to the use of market devices are seldom totally 
valid, because some combination of discharge prices and marketable 
discharge rights can usually be defined which will accomplish what- 
ever the reofulatory system could rea'^onably expect to accomplish. 
MDT\s can be written to require total discharges to be reduced on any 
desired schedule more easily than an equivalent set of discharge 
standards — which must be written in detail for each individual source. 
And they can disallow surges in discharge levels, or require agency 
a]>proval and notification before relocation, or be suspendable under 
certain conditions at least as well as can quantitative discharge 
standards set by administrative processes. If reirional "competition"^ 
for new sources must be discouraged, a uniform discharge price is 
probably more effective and efiicient in doinj?: so than are "uniform" 
discharnre standards, which at best will be different for every industry 
and will usually have to make distinctions among different production 
methods within an industry.* Any distinctions which can reasonably 
be expected from a regulatory system can be accomplished by market 
devices. And as long as the prices or rights are varied only with ex- 
ternnl conditions — i.e., with location or time or discharge characteris- 

* Tn faf't. the dlstlnrtions must sometimes be based on regional differences themselves, 
as when fuel availability or climate vary by region. 


tics, but not with the internal technology or economics of each source — 
then the basic advantages of markets can be maintained.^ 

The fact that market systems can be devised to make (or ignore) the 
same distinctions of time and place which a regulatory system makes 
(or ignores) does not mean that these distinctions should necessarily 
be made (or ignored). The parameters of a market system — the price, 
or the number of MDRs — can be made to vary with external factors 
such as exposed population, flow of the river, average annual sulfur 
dioxide concentration, etc., with relative ease, at least as a technical 
matter; such variation in discharge standards is extremely difficult, 
requiring redetermination and defense of the whole set of complex 
standards in each situation. At the same time, because of their uniform 
application, the parameters of a market system are less easy to vary 
on a case-by-case basis in response to alleged technical problems or 
quiet political pressure. And many of the distinctions made by current 
regulatory programs, particularly those based on simple models of 
what is necessary to accomplish arbitrary ambient standards, are 
probably not worth making in any case. Therefore, market systems 
should be evaluated in terms of whether the distinctions they make 
are reasonable in terms of society's broader interests, not whether they 
are the same ones made by regulatory systems — although they can be 
used to make whatever distinctions are desired. 

Choosing the L&vel of the Price or the Number of Discharge 

The fundamental question in implementing a market system for 
environmental management is: what should the price (or the number 
of marketable discharge rights) be? And the theoretically correct 
answer — that the discharge price, to which polluters will adjust their 
marginal cost of abatement, should be equal to the marginal social 
benefit of discharge reduction — is not much help, since no one knows 
how to estimate the quantities involved. 

One approach to determining the parameters in an environmental 
market system is to begin with ambient standards or desired total dis- 
charge reductions determined by traditional administrative means. 
These could be used directly to define a set of marketable discharge 
rights which would accomplish the objective, or information on con- 
trol costs could be used to estimate a set of prices high enough to ac- 
complish the desired discharge reductions. Either way, the manage- 
ment task should be substantially eased, because the environmental 
agency would not have to define detailed discharge standards for each 
source individually.® * 

In situations where the environmental goals are known and the 
cost of accomplishing them is moderate, the approach based on issu- 
ing MDRs is perfectly reasonable and provides an operational way to 
implement a market system. More difficult but also more interesting 
problems arise when there is no accepted environmental goal, or when 

5 The "delayed compliance penalty" discussed below (p. 329) does not meet this con- 
dition, because the fee is tailored to each source and is applied at the discretion of 
the re^latorv agency. 

« For calculating the price only average control cost information need be used ; in 
setting discharge standards, each source must be dealt with separately. Of course, if an 
ambient standard is so narrowly defined that each source must be controlled individually 
in order to accomplish it. then no real market is possible. In such situations it would be 
wise to reconsider the ambient standard, because such precise source-by-source control is 
seldom necessary for sound environmental reasons. 


the one which exists is so costly to accomplish that it is nonoperational, 
e.^r., the ambient air quality standards in many cities. In these situa- 
tions, choosing the level of the discharge prices is equivalent to decid- 
ing how fast and to what extent society wants to make progress to- 
ward reducing discharges. And, whether regulatory or market proc- 
esses are to be used, this decision is a matter of judgment, to be made 
after consideration of the costs, the benefits, the risks, and the social 
values involved.^ 

Under a market approach to environmental policy, judgments about 
the level of specific discharge prices could be made at the highest 
political levels, in the form of a legislated price. For example, a dis- 
<:*harge price of X dollars per pound could be levied, reflecting the 
legislature's judgment that such a price is a reasonable one to charge 
those who degrade the public's environment, and will provide a useful 
incentive to encourage reductions in discharges. Alternatively, the 
legislature could specify the criteria and considerations which should 
go into the determination of a set of prices by the environmental 
agency, and then exercise general oversight. For example, the agency 
could l>e directed to set prices which would be expected to reduce total 
discharges by half over five years, or which would induce application 
of best available control technolog;V' in most sources,® or which would 
approximate the damages caused by the discharges. The legislature or 
the agency would have to be prepared to defend the prices as "reason- 
nble" in light of the purposes: but with a clear statement of legis- 
lative intent to use discharge prices as an instrument for managing 
society's environmental resources, the courts would be unlikely to 
object to the idea. 

Even with discharge prices set by political and administrative proc- 
esses rather than by technocratic calculation, the market system would 
retain its fundamental advantage: the technical and economic details 
of the how, who and when of discharge reductions are handled by the 
market and not by governmental processes. In choosing the level of 
the price, information on costs, benefits, technology, risks, etc., would 
be weighed, along with basic social values concerning the environment, 
the acceptability of risks, etc. But government would not be deciding 
whether 85 percent or 90 percent control is best for a particular source, 
whether 1975 or 1977 is a technologically feasible deadline, whether 
stratified charge engines are preferable to catalytic converters. And 
it would not be necessary to have periodic crises during which some 
definite standard or deadline must be relaxed legislatively lest the 
economy cease to function.^ The large questions of social values, the 
desirability of going faster or slower, the relative priorities amonrr 
environmental problems can be debated and decided by the social 
processes best suited for these large social questions — the political, 
administrative and legal processes — while the mundane details can be 
left to the social processes best suited for them — the market. 

' Of course, the same is true in situations wliere the "ripht" environmental goals are 
known and the costs are "acceptable." Whether or not it is worthwhile to protect human 
health from all adverse eff(>cts is a matter of jud.crment. as is the definition of a precise 
environmental standard for doinjr so. Thus the dichotomy here is not a clean one. 

" For the purpose of setting the price, information on control costs in different industries 
Would he use<l. but then all would be sub.iect to the ,srrwe price. 

® Tn fact, if it turned out to be more costly or to take loneer to reduce discharges than 
oriirinally thouL'lit. the appropriate policy response might be to raise the price. Tin's possi- 
bility might have some incentive effects on those loolving for solutions to technical 


Evolution of U.S. Environmental Regulation 

Attitudes and policies toward pollution have undergone a steady 
evolution in the United States. In the earliest stages of this evolution- 
ary process, pollution was little more than one of many mechanisms 
by w^hich an individual might impose harm or inconvenience on an- 
other individual or on the public health or welfare, and a harmed 
individual or the relevant public official could always go to court 
and seek redress under civil law. It was generally agreed, however, 
that a polluter had a right to use the air or water for waste disposal, 
so long as he was reasonable in use of it, and the courts would 
weigh the rights of polluter and polluted in reaching a decision. 

Gradually it became clear that some general rules and guidelines 
were necessary to assist the courts in deciding whether the polhiter 
had gone too far in his activities, and anti-pollution statutes began to 
appear. These laws attempted to define certain behavior or results 
of behavior as unacceptable or counter to the public interest, and relied 
on the courts to apply these rather vague definitions to those particular 
sources against which the authorities decided to take action. Eventu- 
ally, the concept of pollution control as a problem in the management 
of social resources became accepted, and progressively more elaborate 
regulatory schemes were developed, culminating in the 1970 Clean 
Air Act and 1972 Federal Water Pollution Control Act. Throughout 
this evolution, however, U.S. policy has continued to treat environ- 
mental problems as purely regulatory problems, with no recognition 
of the possibility that market forces might be useful in managing 
valuable social resources. 

PRE-197 federal LEGISLATION 

As long as pollution was regarded as a problem between identifiable 
polluters and victims of pollution, there was little reason for federal 
action, except where polluters and victims were located in different 
states. Thus, except for a law ^^ regulating oil pollution in tidal waters, 
the federal government was quite slow in enacting legislation to con- 
trol pollution. After a series of efforts in 1936, 1938 and 1940, the first 
federal pollution control law became effective in 1948. 

191^8-65 Efforts 

The Water Pollution Control Act of 1948 left control of intrastate 
waters to the states, and established an extremely cumbersome proce- 
dure for abating interstate pollution. In order to obtain abatement 
under the procedure ^^ the U.S. Surgeon General was required to show 
that pollution originating in one state endangered the health and wel- 
fare of persons in another. He could then give formal notice to the 
alleged polluter and to the relevant state control agency, specifying 
an abatement action and setting a reasonable time for compliance. If 
compliance did not result, he could give a second notice, wait a reason- 
able time, and then establish a federal hearing board which could rec- 
ommend "reasonable and equitable" abatement measures. If these 
recommendations were not complied with, the Federal Security Ad- 

10 The Oil Pollution Act of 1924. 

^ As described in Irwin and Selig, op. cit, pp. 81-83. 


ministrator could, with the state's permission, ask the U.S. Attorney 
General to bring a court action, during which all the issues, from the 
severity of the pollution to the feasibility of abatement, would be con- 
sidered again. Over the years, a series of amendments attempted to 
ex]^and and streamline this process a bit, by extending federal author- 
ity to all navigable waters and allowing federal enforcement in certain 
situations without state approval; and, in 1956, federal funds were 
made available for municipal sewage treatment plants, on a matching 

In air pollution, the federal government was even slower to act. 
Despite the death of some twenty persons during an acute air pollution 
e])isode in Donora, Pennsylvania in 191:8, and the demonstration by 
scientists in the early 1950s that Los Angeles smog was caused by auto 
emissions, it was not until 1955 that a federal air pollution control act enacted, and then only established a small program of research, 
training and demonstrations. 

The Clean Air Act of February, 1963 (P.L. 84-159), was the first 
air (juality legislation giving the federal government enforcement 
powers, and was modeled after the earlier water pollution control 
law : the Secretary of HEW could call an abatement conference when 
interstate pollution was harming public health and welfare, or when 
invited to by a state; the conference could be followed by hearings 
and. hnally, federal court action. In 1965 a second title (P.L. 89-272) 
was added to the Act, authorizing the Secretary of HEW to set 
•emission standards for automobiles as soon as practicable, giving "ap- 
inopriate consideration to technological feasibility and economic 
costs :'' the federal standards adopted under this provision for applica- 
tion in 1968 were about the same as the state standards California had 
imposed for 1966. 

1965-70 Strategy 

By 1965, it was clear that these early federal ventures into pollution 
control were seriously defective. Thus, the 1965 Water Quality Act 
and the 1967 Air Quality Act ^^ adopted a rather different approach. 
Tender these Acts, the Secretary of HEW was to produce documents 
(lescril)ing what was known about the effects of air and water pollution 
on pt'0])le, other organisms, materials, etc. Each state was then to es- 
tablish ambient (i.e., "surrounding") air and water quality standards 
for the sev^eral air quality control regions and water bodies within the 
state, based on the intended uses of the region and HEW's description 
of the characteristics of air and water which were consistent with these 
intended uses. These ambient standards would be stated in terms of 
measurable quantities, e.g., micrograms per cubic meter of sulfur di- 
oxide in the air, temperature and oxygen content of the water, number 
'Of bacteria per milliliter, etc., and, once approved by HEW, would 
not have to be reconsidered every time an action was taken against a 

The next stage in the new strategy was the translation of these am- 
bient standards into specific effluent and emission levels for individual 

'2 Public Law 89-234 and Public Law 90-148, respectively. Although the Acts were ad- 
ministered by different agencies and were different In details and terminology, they were 
quite similar in basic structure and logic, and are described here in a single, somewhat 
i^tylized form. 


sources. Each state was required to develop implementation plans de- 
scribing the effluent and emission limitations it would enforce and 
other st^ps it would take to accomplish the ambient standards in each 
region and basin. If any state failed to produce ambient standards or 
implementation plans which were acceptable to the Secretary of HEW, 
he could promulgate his own. Once a plan had been promulgated, the 
Secretary could take enforcement actions against individual polluters 
if any state failed to do so. The 1967 Air Quality Act also provided 
research funds for development of control technologj^, and required 
the Secretary of HEW to set emission standards for new automobiles, 
"giving appropriate consideration to technological feasibility and eco- 
nomic costs." ^^ The 1965 Water Quality Act also continued the policy 
begun in 1956 of- granting federal funds on a matching basis for the 
construction om municipal sewage treatment plants. 

From a legal and regulatory perspective, the strategy adopted in the 
mid-1960s was quite a reasonable one. The general goal of "enhancing 
and maintaining" the quality of the nation's air and water resources 
would be translated into specific, physically-definable and measurable 
characteristics, which would become established parts of the law after 
the required public hearings, administrative processes, and judicial 
review. These would then be translated into requirements on individual 
pollutant dischargers, also stated in specific, measurable terms, and 
presumably defensible against challenge on the grounds that they were 
necessary to accomplish the ambient standards. Enforcement should 
be relatively easy. Unfortunately, it did not work out that way. 

For air pollution, the process hardly worked at all.^* By 1970, tlie 
National Air Pollution Control Administration (NAPCA) had not 
defined the air quality control regions, and had published the "criteria" 
documents describing the effects of air pollutants for only two pollu- 
tants. Not a a single state had a complete set of ambient standards and 
implementation plans for a single pollutant, and the plans which did 
exist had not been and likely would never be reviewed for adequacy 
by NAPCA. No enforcement actions were ever taken under the proce- 
dures of the 1967 Act. And for mobile source emissions, the standard- 
setting process did not hold great promise of accomplishing the needed 
emission reductions. 

For water pollution, the only part of the process which w^orked rea- 
sonably well was the establishment of water quality standards, i.e., 
the goals, in terms of physical and chemical characteristics of water 
bodies. Most states had set reasonable water quality standards by 1971.^^ 
But the process began to break down at the level of translating these 
goals into enforceable regulations, as deciding what total discharge 
levels were consistent with the desired water quality and which dis- 
chargers should be allowed how much of this total proved too much 
for the states. Following "guidelines" put out by the Federal Water 
Pollution Control Administration (FWPCA), most of the states 
adopted, and the FWPCA accepted as adequate, general and often 
vague requirements for secondary treatment (or equivalent) and "non- 
13 Public Law 89-272, Section 202(a). 

" See discussions In : John Esposito, Vanishinff Air, Center for the Study of Responsive 
Law, 1970, pp. 158-172 ; Alan Kneese and Charles Schultze, Pollution, Prices, and Public 
Policy, the Brookings Institution, 1975, pp. 48-50; GAO Report, "Assessment of FederaL 
and state Enforcement Efforts," 1973, p. 35. 

15 See N. W. Hines, "Public Regulation of Water Quality In the United States," report 
to the National Water Commission, 1971, pp. 254-255. 


de<2:radation-' of existing hifrli-quality waters, usually- with no time- 
tables or specific actions required. The enforcement process remained 
unworkable — even thoui^h it was no lon^rer necessary to prove that a 
specific discharge endangered the public health and welfare, it was 
still necessary to prove that a specific discharcre was the cause of a vio- 
lation of the water quality limitation. With notable exceptions, the 
states did not vigorously enforce, and federal enforcement was 
limited to cases whore interstate harm could bo demonstrated and a 
state invited the federal government to take action. Thus, progress 
under the 1964 Water Qualitv Act was disappointing. Even the mu- 
nicipal construction grants program was having disappointinir results, 
as the federal funds wore allocated carelessly and inefficiently.^^ 

As frustration with progress was growing in the last lOnOs. the pos- 
sibilities offered by the Refuse Act of 1899 were discovered.^^ This old 
law. which had never been used to control pollution, forbids the dis- 
cliarge of virtually anything into navigable waters, except that the 
Secretary of War '*may permit the deposit of any material . . . within 
limits to bo defined and under conditions to bo proscribed by him . . . ; 
and whenever any permit is so granted the conditions thereof shall be 
strictly complied with, and any violation thereof shall be unlawful.'' ^^ 
Thus, undo]- the Act. tlio Secretary of War (Defense), acting throufrh 
the Army Corps of Engineers, could prohibit any industrial ^^ dis- 
charge which had not been granted a permit, and could put whatever 
conditions he chose on any permit he chose to issue. 

Tlie Refuse Act seemed to grant to the federal government all the 
authority it needed to control water pollution. In December 1970. Pres- 
ident Xixon. by Executive Order, directed the Army Corps of Engi- 
neers and EPA to issue guidelines for permits under the Act, and some 
court cases resulted in convictions of discharges for failure to obtain 
the needed permits. Quickly, however, the courts ruled that the ad- 
ministrator of any permit program under the Act could not simply 
grant or withhold a permit as he saw fit and must, in fact, prepare an 
environmental impact statement -° on every major action — the Refuse 
Act permit program died aborning. But. even in its brief lifetime, the 
advantages inherent in a permit system, in which operating without a 
permit is unlawful and the permit can be used to imj^ose specific con- 
ditions, had been demonstrated. 

In 1909 and 1970, "the ecology" became an important issue on the 
American political scene. There was widespread dissatisfaction with 
the way environmental matters were being disregarded in important 
decisions and with the ineffectiveness of those programs of regulation 
which had been established. In 1969, the National Environmental 
Policy Act was enacted, establishing the Council on Environmental 

I''' See Report to Congress. "Examination Into tlie Effectiveness of the Construction Grant 
Program." U.S. GAO. Nov. 3. l!*f>9. 

- Primarily by Representative Henry Reuss and his staff on the Subcommittee on Con- 
sorvation and Natural Resources, House Committee on Government Operations, See Irwin 
and Selifr, op cit. p. 67. 

'* Cited in R. Zener. "The Federal Law of Water Pollution Control," In Federal Environ- 
wentnl Law. Environmental Law Institute, 1974. p. 785. 

1^ The courts had held that municipal discharees were not covered. 

^ The EIS is a procedural requirement of the National Environmental Policy Act. See 
text, p. 276 below. 


Quality as an advisory body to the President and declaring it to be 
national policy that the government shall endeavor "to create and 
maintain conditions under which man and nature can exist in produc- 
tive harmony." ^^ NEPA's most important substantive requirement was 
that federal agencies must prepare a detailed statement describing 
the environmental impact of any proposed action which might have 
significant effect on environmental quality, and must study reasonable 
alternatives to the proposed action.^^ The Clean Air Amendments 
(CAA) of 1970 and the Federal Water Pollution Control Amend- 
ments (FWPCA) of 1972 were enacted to expand the federal role in 
environmental regulatory policy and, in slightly amended form, pro- 
vide the basis of the current programs. 

These two pieces of legislation built on the earlier pollution control 
programs and, especially the CAA, kept many of the then-existing 
program features. But there are so many new provisions and regula- 
tory innovations in these laws that they are best regarded as initiating 
a new era in environmental regulation. In this section the principal 
regulatory provisions of the 1970's legislation are described and then 
some general observations on the new policy directions are made. 

The 1970 Clean Air Act 

The 1970 CAA kept the overall structure of the 1967 legislation, 
which set emission standards for new automobiles and required the 
states to develop implementation plans describing how ambient air 
quality standards would be accomplished. The changes introduced 
into this process by the 1970 Act were, however, profound : the Con- 
gress itself set the levels and deadlines for the automobile standards ; 
EPA was directed to establish, on the basis of narrow considerations, 
national ambient standards within months; the states were required 
to develop and enforce plans to meet these ambient standards on strict 
timetables, or have EPA do so for them ; and citizen groups were given 
standing in court to sue the Administrator if EPA failed to act as 
directed. The 1970 Act also introduced, for the first time, national 
emission standards for certain stationary sources and contained lan- 
guage which the courts interpreted to require a policy of "non-de- 
terioration" of clean air areas. 

Federal Mobile Source Emission Standards 
The mobile source emission standards set by HEW under the earlier 
legislation had not produced much technological progress by the 
automobile industry and total auto emissions were projected to in- 
crease under proposed standards as the vehicle population grew in 
the 1980s. Despairing of the strategy ^^ which allowed administrators 
discretion in setting standards, "giving appropriate consideration to 
technological feasibility and economic costs," Congress wrote into the 
1970 CAA specific numerical standards, requiring a 90 percent reduc- 
tion in vehicle emission levels within five years.^* The Administrator 

21 Section 101(a) of NEPA, Public Law 91-190 (1970). 

^ Since NEPA Is not a regulatory program, it will not be dealt with here. For a good 
description of how NEPA was implemented in its early years, see F.R. Anderson, Jr., 
"The National Environmental Policy Act," in Federal Environment Law, 1974. 

23 See John F. Bonine. "The Evolution of 'Technolo?:y-Forcing' in the Clean Air Act," 
Environment Reporter, Monograph No. 21, July 1975, for a good discussion of the evolu- 
tion of Congressional attitudes. 

2* Hydrocarbons (HC) and carbon monoxide (CO) were to be reduced to 10 percent of the 
1970 standards by 1975, and nitrogen oxides (NOX. which were uncontrolled in 1970) to 
10 percent of their average measured level in 1971 by 1976. 1970 CAA. Sec. 202. 


was allowed to extend the compliance dates only by one year, one 
time, and then only if he were convinced that the auto manufacturers 
were making good faith efforts to comply but could not for technical 
reasons, and that extension was in the public interest. For enforce- 
ment, EPA was to issue regulations concerning testing, fuels, war- 
ranties, etc. Violations were subject to a $10,000 fine per vehicle, 
sufficient to shut down the industry. 

h\ setting these standards and these dates. Congress was not being 
totally arbitrary: scientists of HEWs National Air Pollution Con- 
trol Administration had argued that 90 percent reductions in emis- 
sions would be necessary^ by 1980 to be assured of protecting the 
public health in the worst cities." But in choosing these specific 
numbers and dates. Congress was explicitly not trying to strike a 
balance betw^een costs and benefits. In keeping with the spirit of the 
times, the objective was total elimination of the adverse effects of 
pollution, whatever the adverse effects of doing so might be. As 
Senator Muskie said during the Senate debate on the Amendments; 
"The deadline [for automobile emission standards] is based not, 
I repeat, on economic and technological feasibility, but on considera- 
tions of public health." ^^ The automobile industry would clean up 
or be shut down — unless, of course, they could convince Congress to 
change its mind. 

National Amhlent Air Quality Standards 

Under the 1970 CAA, the Administrator of EPA has the responsi- 
bility of dividing the country into Air Quality Control Regions 
(AQCR), publishing a list of air pollutants, issuing for each pollutant 
a criteria docinnent which describes the effects of the pollutant "in the 
ambient air" on the public health and welfare, and then establishing 
National Ambient Air Quality Standards (NAAQS). Rigid dead- 
lines are included, forcing the Administrator to act within months of 
the date of enactment. ^^ 

The significant innovations here are the introduction of national 
ambient standards and the fact that the "primary" NAAQS, which 
are the principal action-forcing goals of this part of the Act, are to 
bo set strictly on the basis of the scientific evidence contained in the 
criteria document: the Administrator is to use his best judgment to 
set primary NAAQS at such levels which "are requisite to protect the 
public health . . . allowing an adequate margin of safety." ^^ The 
Act and the legislative history leave little doubt that in setting the 
primary NAAQS, considerations of the technical and economic feasi- 
bility of accomplishing them should play no role — Congress estab- 
lished the goal of having the air so clean that even "particularly 
sensitive citizens such as bronchial asthmatics and emphysematics" ^^ 

--''• D. Barth. "Federal Motor Vehicle Emission Goals," Journal of the Air Pollution 
Association, 1970. 

-"« Senate Keport No. 91-1196, 91st Congress, 2d Session, 1970, p. 32905. Cited in Bonine, 
op. cit., p. 19. 

^ For example, AQCRs had to he designated within 90 days, and proposed NAAQS had 
to be published within 80 days and promulgated within 90 days after such publication, for 
thiisp six pollutants for which criteria documents had eventually been issued under the 
ion: Act. 

2« Clean Air Act. Sec. 109(b) (1). 

-•' Sonato rommitte on I»;;l)lic Works, in S. Rep. No. 91-1196. 91st Cong., 2nd Siess.. 20 
COTO), cited in .Torling. "The Federal Law of Air Pollution Control," in Federal Enri- 
rojwiental Laiv, ELI, 1974. p. 1082. The Committee goes on to say that In protecting the 
health of such people, "reference should be made to a representative sample . . . rather 
than to a single person." 

8.-^-944—78 19 


would suffer no adverse effects from air pollution at any time, any- 
where in the country. "Secondary" NAAQS, "requisite to protect 
the public welfare from any known or anticipated adverse effects 
associated with the presence of such air pollutant in the ambient air," 
are also called for ; whether economic and technical feasibility should 
enter into the setting of the secondary NAAQS remains unclear. 

Btate iTYiplementation Plans 

The CAA requires each state to adopt and submit to the Adminis- 
trator, within nine months after promulgation of a primary NAAQS, 
an implementation plan for each Air Quality Control Region within 
the state. The state implementation plan (SIP), "shall" be approved 
by the Administrator if it meets certain conditions, the most impor- 
tant being that it contain specific "emission limitations, schedules, 
and timetables for compliance with such limitations, and such other 
measures as may be necessary to insure attainment and maintenance 
of [the primary NAAQS by 1975] including, but not limited to, land- 
use and transportation controls." ^^ There is nothing in the SIP 
approval criteria regarding the social, economic or technical feasibil- 
ity of accomplishing the primary NAAQS, except that the Adminis- 
trator may delay the compliance date by up to two years, once. If a 
state fails to submit an acceptable SIP, i.e., one which specifies 
exactly how the state will, at any cost, accomplish NAAQS by 1975 
or 1977, EPA must itself promulgate such a plan for that state. 

The drafters of the 1970 CAA had explicitly recognized that attain- 
ment of NAAQS would require the SIPs to include land-use and 
transportation controls — although they might not have realized just 
how severe these would have to be. Court decisions ^^ soon after enact- 
ment extended the Act's influence on local transportation even further. 
The Act's objective of "attainment and maintenance" ^^ of air quality 
was interpreted to require the SIPs to contain controls which would 
prevent future growth .from causing a future violation of NAAQS, 
once attained. Thus, EPA was given the authority to require the states 
to develop air quality maintenance plans and programs which would 
project future growth, identify potential violations of NAAQS, and 
take steps to prevent these violations. In addition, EPA promul- 
gfated regulations requiring preconstruction review and control of new 
"complex" or "indirect" sources, i.e., a facility such as a shopping 
center or sports arena which might attract enough motor vehicles to 
cause a local violation of NAAQS. Air quality maintenance and in- 
direct source controls taken together constituted an apparently power- 
ful tool for controlling local land use and transportation at the federal 
level, in the name of air quality.^^ 

National Emission Standards for Stationary Sources 

^ Before 1970, national emission standards had not been seriously con- 
sidered, except for mobile sources, where there are clearly special cir- 
cumstances. And the 1970 CAA had as its principal mechanism for 

30 3 970 CA.A. Section 110(a)(2)(B). 

31 Particularity NRDC v. EPA, 475 F. 2d 968, 4 ERC 1945 (D.C. Cir., 1973). 

32 CAA. Sec. lt9(a)(2)(R). 

"3 See diwciis^ion in L. .T. Dnnielson. "Control of Complex Emission Sources — A Step 
Towards Land Use Planning," in Ecology Law Quarterly, vol. 4. No. 3 (1975). 


controlling stationary sources the ISTAAQS and state emission regula- 
tions. But by 1970, some persuasive arguments were being made for 
national stationary source emission standards. In particular, it was 
feared that the strategy based on ambient standards would encourage 
sources to locate in clean regions, giving these areas a competitive eco- 
nomic advantage over others and causing the further spread of pollu- 
tion. It was also realized that there may be some harmful pollutants 
for which a NAAQS had not yet been or could not be promulgated 
for technical reasons, and some mechanism was required for controll- 
ing these. So the 1970 CAA provided for national emission standards 
for new sources and for hazardous pollutants. 

Section 111 of the Clean Air Act requires the Administrator of 
EPA to publish a list of "categories of stationary sources" which, in 
his determination, "contribute significantly to air pollution which 
causes or contributes to the endangerment of public health or welfare." 
He must then establish "standards of performance" for new sources in 
these categories, which reflect "the best system of emission reduction 
which (taking into account the cost . . .) the Administrator determines 
has been adequo.tely demonstrated." The Administrator is also directed 
to extend these techuology-based emission limitations to existing sour- 
ces which are not otherwise controlled. Thus, after a simple determina- 
tion that a category of s(nirces "may contribute substantially to air 
pollution," and a determination that a specific emission standard re- 
flects the best demonstrated system, "it shall be unlawful . . .to operate 
... in violation" of that standard, and EPA can take direct enforce- 
ment action against any violator. 

Similarly, Section 112 requires the Administrator to publish a list of 
, "hazardous air pollutants," meaning pollutants which may "contrib- 
I ute to an increase in mortality or an increase in serious irreversible, or 
' incapacitating reversable illness," but for which there is no national 
ambient air quality standard. He must then establish for each hazard- 
ous pollutant an "emission standard ... at the level which in his 
judgment provides an ample margin of safety to protect the public 
health . . ." Xo new source may begin operation in violation of this 
, emission standard, and existing sources have no more than two years 
I to comply or shut down. The economic or technical feasibility of com- 
pliance is not to be a factor, except that the President of the United 
States may exempt any stationary source from compliance for renew- 
able two-year periods if he finds that technology for compliance is 
unavailable and continued operation of the source "is required for 
reasons of national security." Thus, the Administrator need only de- 
: tennine that, "in his judgment," some specific emission level is neces- 
sary to provide "an ample margin of safety to protect the public 
health," and all sources unable to comply or get a Presidential waiver 
must close down in two years. 

Nonde terioration 

The CAA has as a purpose "to protect and enhance the quality of 

tlie Nation's air resources," ^* a rather ambiguous statement. Since the 

secondary NAAQs are supposed to eliminate all known or anticipated 

adverse effects of air pollution, if they are "properly" chosen it would 

»*CAA, Sec. 101, (b)(1), emphasis added. 


seem that getting and keeping the air as clean as the secondary 
KAAQS would accomplish about all the protecting and enhancing 
anybody could reasonably want. This interpretation is strengthened 
by the fact that the Administrator is directed to approve any SIP 
which provides for the attainment and maintenance of the NAAQS. 
Nonetheless, the courts quickly ruled that the reference to "enhance,'' 
when combined with the legislative history, required EPA to pre- 
vent ^'significant deterioration" of existing air quality in any portion 
of any state.^^ Although the precise meaning and implications of this 
are still being debated, it became clear early in the history of the 
CAA that the policies to control air quality could be an important 
influence on the location of future national economic growth. 

The 1972 Fedeml Water Pollution Control Act 

The 1972 FWPCA maintained most of the provisions of the earlier 
legislation, but only in an effort not to disrupt those stat-e programs 
which were accomplishing something. On the whole, the 1972 act 
took an^ entirely different philosophical tack, based on the uniform 
application of the best technology everywhere. To accomplish this it 
set ambitious goals and rigid deadlines, established complex bureau- 
cratic processes, promised billions of dollars to pay for the bulk of the 
municipal cost of accomplishing the goal, and provided for "citizen 
suits'' to force action. 

The Ohjectwes and Timetahles 
The 1972 FWPCA states two "national goals :" "that the discharge 
of pollutants into the navigable waters be eliminated by 1984;" and, 
"wherever attainable, an interim goal of water quality which pro- 
vides for the protection and propagation of fish, shellfish, and wild- 
life and provides for recreation in and on the water be achieved by 
1 July 1983." ^^ These ambitious national goals are to be pursued 
with a technology-based strategy. Section 301 of the Act requires that 
all point source discharges be using "best practicable control tech- 
nology currently available" (BPT) by 1 July 1977, and "best avail- 
able technology economically achievable" (BAT)^^ by 1 July 1983. 
Other sections of the Act require national effluent standards for new 
sources (Sec. 309), for sources which discharge to publicly owned 
treatment plants ("pretreatment" standards. Sec. 307(b)), and for 
sources of toxic pollutants (Sec. 307(a)). In defining these effluent 
standards, EPA is to require levels attainable with "best" technology 
always, and zero discharge wherever possible. Thus, the 1972 FWPCA 
"clearly establish (es) that no one has the right to pollute — that pol- 
lution continues because of technological limits, not because of any 
inherent right to use the nation's waterways for the purpose of dis- 
posing of wastes." ^^ The Act's objectives and deadlines are to be 
studied by a special National Commission on Water Quality and per- 
haps modified by Congress; but the Administrator has no authority to 
relax them. 

"^ Sierra Cluh v. RiicTcelshaus, 344 F. Supp. 253, and Fri v. Sierra Cluh, 412 U.S. 541. 

36 1972 FWPCA. Section 101(a). 

37 Or its ponivalent. The FWPCA specifies performance standards, to be met any way 
the discharper chooses. ^ ^ - ^ ,, ^n 

^^ ?;pnntp Report on 1972 Amendments, S. Rep. No. 92-414; 92d Congress, 1st Sess., 42 
1971). Citpd in Zener. op. cit., p. 694. 

. 281 

National Pollution DiscJutrge Elimination System 
Under the 1972 FW'PCA, EPA is required, with short deadlines, to 
begin producino: the technical and economic information needed to 
establish all the effluent standards envisioned. "Effluent ^udelines** are 
to be produced, to be used in establishing BPT and BAT effluent limi- 
tations for individual existing sources. In addition, pretreatment 
standards and toxic effluent standards are to be promulgated. 

To translate these guidelines and standards into regulations en- 
forceable on individual dischargers, the 1972 FWPCA introduced a 
permit system called tlie Xational Pollutant Discharge Elimination 
System (NPDES). This system required EPA (or the state, where 
a state has an EPA-approved program) to write a permit for each 
individual source, specifying (among other things) how and on what 
schedule the applicable effluent standards would be accomplished, in no 
case later than 1 July 1977 for BPT on existing sources. Where the 
national effluent standards were not sufficient to accomplish state water 
quality standards, more stringent requirements were to be imposed, 
bavsed on an analysis of effluent limitations necessary to accomplish 
water quality standards. No point source could discharge without a 
permit. Once the permit was in place, the state, EPA or interested citi- 
zens could go to court to compel compliance with its conditions. 

Municipal Pollution Control 

The 1972 FWPCA makes municipal point source dischargers sub- 
ject to the XPDES permit requirements, the same as industrial dis- 
chargers. EPA was to pronmlgate effluent guidelines defining "second- 
ary treatment or its equivalent'' to be met by 1977, and "best practi- 
cable waste treatment teclmology over the life of the works'' to be 
met by 1983, and tlien to issue permits specifying how individual 
municipal plants will come into compliance on time. It was obvious, 
however, that federal enforcement action against municipalities is not 
a realistic policy so long as the municipalities could claim that they 
could not afford treatment, and the FWPCA relies primarily on mas- 
sive federal construction grants to bring about control of municipal 

To see that these federal funds are well spent. Section 208 of the 
FWPCA requires states to develop areawide waste treatment plans 
as the basic long-term management plans for each area. These plans 
are to identify the best regional waste treatment systems, set priorities 
for construction, and, together with the facilities plans required by 
EPA regulations, find cost-effective solutions to areawide problems. 
The Section 208 plans are also to identify the principal "non-point" 
source ^^ problems and develop strategies for dealing with these. 

EPA's "1971 Needs Survey" had estimated that $18.2 billion of 
federal funds silent during fiscal years 1972 to 197G, on a 3 to 1 match- 
ing basis, would bring about significant progress toward the 1977 
"secondary treatment" goal.*" Hence, the 1972 FWPCA authorized 
obligations of $5 billion in fiscal year 1973, $6 billion in fiscal year 
197-1, and $7 billion in fiscal year 1975, recognizing that this was prob- 
nMy inadequate to support full compliance with the 1977 requirement, 

s-' "Non-point" sourops include agrricnltural lands, forests, city streets, etc. 
*^ Draft Staff Report, ^CWQ, p. V-36. 


but expecting that, carefully planned and managed, this much money 
could accomplish much by 1977. President Nixon explained his veto 
of the 1972 FWPCA in terms of the "inflationary impact" of this $18 
billion; but apparently did not persuade Congress; the veto was 
promptly overridden. 


The new, post-1970 federal environmental policies were w^idely 
hailed as dramatically different approaches to the management of 
environmental resources. And, indeed, in their extension of federal 
autliority, "economics-be-damnecl" attitude, use of rigid goals and 
deadlines backed up by citizen standing in court to force action, and 
level of financial commitment they were quantitatively different. In 
the more qualitative dimensions, however, the new policies were noth- 
ing but a logical extension of the old: they continued to assume that 
environmental problems would be so simple and easy to solve that ^ 
traditional administrative mechanisms would be adequate to the task 
with no help from market forces. 

The drafters of the 1970 CAxV and the 1972 FWPCA were clearly 
more interested in overcoming the traditional obstacles to effective 
environmental regulation than in accommodating the complex facts 
of reality or the broader needs of society. National ambient air quality 
standards would be established, and on the basis of the Administra- 
tor's judgment about the limited scientific information concerning a 
jiarrow range of effects, because this would speed up the regulatory 
process and minimize the opportunities for challenge, not because the 
effects of air pollution are so simple and well understood that a simple 
numerical standard is an adequate statement of society's environ- 
mental goals. National discharge standards would be used because 
(it was felt) this would allow a central bureaucracy to make a sina^le 
set of narrowly-based determinations rather than requiring case-by- 
case determinations involving complex trade-offs. State plans for ac- 
complishing ambient goals would be produced too quickly for careful 
thought and then would be legally enforceable, because this would 
short-circuit many difficult questions about technology, economics, 
and environmental effects. Rigid deadlines would be set, and citizens 
given the right to force the Administrator to act, because otherwise the 
process would bog down interminably, not because nature promised 
doom by 1975 or iDecause society was well-served by rushing forward. 
In short, the needs of the economy and of the environment for flexi- 
bility, discrimination, and careful timing were subordinated to the 
needs of regulatory processes for rigidity, uniformity and a sense of 

Perhaps in the early 1970s it was possible to believe that solving 
environmental problems would be a simple, quick and inexpensi^^e 
task, if only enous-h political will were applied to it. There is, after all, 
no reason to establish costly and sophisticated management systems 
for problems which cnn be solved cheaply and simply. But. as the ex- 
perience reviewed in the next section reveals (and as should have been 
realized even in 1970), environmental problems are not going to be 
quick, simple and cheap to solve, and the management concepts im- 


plicit in the environmental policies of the earl}^ 1970s are not likely 
to prove adequate in the long run. 

Evaluation of the 1970 Ci^ean Air Act 

Evaluation of a program as ambitious, complex, and unprecedented 
as that launched by the 1970 CAA is extremely difficult and cannot 
be expected to produce unambiguous conchisions. The easiest observa- 
tion to make is that the literal goals of the Act have not been met or 
are unlikely to be met : national ambient air quality standards were 
not achieved everywhere by 1975 or by 1977, the latest fallback date 
envisioned in the 1970 CAxV ; automobile emissions were not reduced 
90 percent by 1975 and will not be by 1980. In the sense of strictly ac- 
complishing its literal goals, the federal air quality management pro- 
gram has clearly "failed." 

This criterion, however, is neither fair to the drafters of the lesfis- 
lation nor useful to those trying to improve it. The specific goals and 
deadlines in the 1970 CAA — indeed, any very precise, quantitative 
statem.ont of goals and timetables — do not reflect the needs of society 
and nature as much as they reflect the needs of administrative and 
legal processes, especially within a purely regulatory system, and one 
cannot criticize a program purely for failure to accomplish goals 
which might have served their administrative purposes without being 
accomplished. Instead, one must look at what the programs have 
actually accomplished as they have been amended over the years, make 
some judgment about what they are lilcely to accomplish in the future, 
and then compare this performance with what one could reasonably 
expect from any regulatoiy program or some non-regulatory 


The principal regulatorv innovations of the 1970 Clean Air Amend- 
ments were the use of legislatively-set, uncompromisable. health-pro- 
tection goals, and rigid deadlines to force action. And the principal 
fact of life since 1970 has been the periodic postponement of deadlines 
and revisions of timetables by Congress, the courts, and EPA. This ob- 
servation by itself does not demonstrate that the CAA was necessarily 
unwise in setting these rigid deadlines; and it certainly does not sug- 
gest that it was unwise to relax the deadlines as it has become clear they 
would not or should not be met. But the experience so far with the 
CAA does suggest that setting rigid deadlines and constraining the 
reo-ulatory authorities do not do much to overcome the basic problems 
with regulation in this area. 

J/nhile jSource Emission Standards 

The 1070 Amendments required that emissions of pollutants from 
new automobiles (on a gram per mile basis) should be reduced by 
ninety percent within five years. The Administrator was allowed to 
grant only a single, one-year extension, and then only if he found the 
manufacturers were making serious efforts but could not comply for 
technical reasons. The burden was to be on the manufacturers to find 
a solution and to bear the costs. 


Immediately, the manufacturers did the rational thing under the 
circumstances : they set out to prove it was impossible to meet the 
standards, and applied for an extension of the 1975 HC and CO stand- 
ards. In May 1972, EPA Administrator Kuckelshaus rejected the appli- 
cation for extension, on the grounds that the manufacturers had failed 
to prove the standards could not be met, and suggested that the most 
promising method of meeting the standards was the catalytic converter. 

The automobile manufacturers then took the issue to court. In Inter- 
national Harvester v. Ruckelshaus^ the court remanded the issue back 
to EPA, on the grounds that EPA had not adequately refuted the man- 
ufacturers' claim of technical infeasibility, and that the economic con- 
sequences would be disastrous if events proved EPA to be wrong. EPA 
granted the extension for HC and CO, setting an interim standard for 
1975 automobiles, and three months later granted a one-year extension 
for the 1976 NOX standard.^^ 

EPA and the courts had now done all they legally could to relax the 
deadlines, and the battleground shifted to the political arena. The Arab 
oil embargo and the "energy crisis" provided a convenient opportunity 
for statesmanlike reconsideration of those earlier decisions made at 
the height of the "environmental crisis," and the Energy Supply and 
Environmental Coordination Act of 1974 (ESECA) was the result. 
Among other things, ESECA moved back by two years the original 
deadlines for accomplishing the 90 percent emission reductions (from 
1975 to 1977 for HC and CO, from 1976 to 1978 for NOX) , and allowed 
the Administrator of EPA to grant a further one-year delay for HC 
and CO; this he proceeded to do in March 1975. In 1977, Congress 
amended the CAA again, postponing the emission reductions two 
more years. 

As a result of the federal regulatory program under the 1970 CAA, 
the emission levels from new automobiles have declined, but much more 
slowly than originally contemplated by the 1970 Act, as illustrated in 
the table below : 

Table 1. — New automoMle emission char act eristics * 




Actual 1970 emission levels 

1970 CAA standards for 1975-76 « 

Actual 1975 standards. 


....: .41 






7. 0-3. 4 




Actual 1978-79 standards 



1977 amendments standards for 1980-81 » 



1 All numbers are in grams per mile. 

2 To be met in 1975 for PIC and CO, in 1976 for NOX. 

* To be met in 1980 for HC, in 1981 for NOX; for CO, standard is 7.0 in 1980, 3.4 in 1981. 

The above table indicates that emission levels in 1979 will probably 
be about four times what they were supposed to be in 1975 under the 
original 1970 CAA, and will have been reduced only about 60 percent 
from the actual 1970 levels. If no further delays are encountered, the 
90 percent reduction in emission levels which was 4-5 years away in 
late 1970 is 2-3 years in the future now (in 1978) . 

« See discussions In J. T. O'Connor, "The Automobile Controversy," and C. M. Ditlow, 
"Federal Regulation of Motor Vehicle Emissions," in Ecology Law Quarterly, vol. 4. No. 3 


Is a 60-percent reduction in new automobile emission character- 
istics between 1970 and 1979 evidence of success or failure of the 1970 
C'AA strategy? Opinions may differ on this question, especially since 
tlie discovery that the NOX problem may not be as severe as had been 
thought/^ the energy crisis which motivated ESECA, and the dis- 
covery that the catalytic converter could produce sulfuric acid emis- 
sions were just the sort of developments which would cause changes 
in any policy. Although the original objectives have not been strictly 
met, it may be that progress has been better than it would have been 
under a more "reasonable" policy, i.e., one which set less demanding 
(leadlines or allowed the Administrator more discretion. 

On the other hand, by focusing all attention on the catalyst and an 

aibitrary, rigid, but ultimately negotiable deadline, the CAA strategy 

.has neither produced dramatic shortrun improvements nor laid the 

i:!Ouudwork for continuing progress in the long run. xVs discussed 

-cwhere in this chapter, regulation is not particularly effective in 

limulating ingenuity and continuing technological improvement. 

National Ambient Standards and State Implementation Plans 
The principal implementation device of the CAA is the use of fed- 
crallv-set national ambient air quality standards (XAAQS), and state 
plans for meeting these. The State plans are to contain measures for 
controlling emissions from existing stationary sources, and for con- 
trolling transportation and land use, to the extent necessary to ac- 
complish the NAAQS. 

National Ambient Air Quality Staiulards 

The NAAQS for the six pollutants explicitly mentioned in the 1970 
amendments were promulgated as required : in 1976, plans were an- 
nouuced for a seventh "criteria" pollutant, lead. There has been only 
one legal challenge to the XAAQS, even though that one resulted in 
EPA relaxing a standard/'' And an independent scientific review of 
the XAAQS by the Xational Academy of Science has "found no 
substantial basis for changing the standards." ** Since the CAA gives 
great discretion to the Administrator in choosing (primary) standards 
which protect public health with an adequate margin of safety, since 
the scientific evidence is poor, and since Administrator Ruckelshaus 
was careful to err on the side of overprotection in setting the XAAQS 
initially, it is not surprising the coui'ts and the scientific establishment 
are reluctant to question his judgment. 

Perhaps a more important reason for the lack of legal challenge to 
the XAAQS is that it quickly became apparent to most participants in 
the struggles over air pollution control that, despite the intent of the 
1970 Amendments that health protection should drive the program, 
the real issues would continue to be technology and economics. The 

"-In 1972. it WIS disoovprpd that thp approvod method for moasurint? nitro'^en o^ide 
actually overestimated ambient levels and. as a result, the severity of the NOX problem 
had bpen overstated. See discussion In 1074 CEQ Annual Report, p. 274. 

^' Whfti nskofl bv tbr. ronrt simplv to explain Imw the nnnnnl-nverai''^ concentraHon Tmit 
of the second sulfur dioxide NAAQS related to Information in the criteria document. EPA 
choso to withdraw this part of the standard, leaving a three-hour limitation. Kennecott 
Copper V. EPA. D.C. Circuit (1073). 

<4 National Academy of Soionces National Academy of Ensrlneerincr. Air Ounlifn and 
Aiitorrtohile Emission Control, Vol. T. Snmmarv Report prepared for the U.S. Senate Com- 
mi^'.-lon on Public Work« Sec. No. 93-24 (Wash., D.C. USGPO, 1074), p. 6 Cited In 1975 
CEQ Annual Report, p. 46. 


"threshold theory" *^ on which the concept of NAAQS was based has 
never had firm scientific support, but was thought to be necessary for 
an effective regulatory program. Once the concept was in the law and 
EPA had picked the numbers, the EPA research apparatus went to 
work, to demonstrate that, in tact, there were thresholds at the chosen 
level.^^ P)Ut as it has become clear that the costs of meetinjj the chosen 
NAAQS would be prohibitive in some cases, and that doing so would 
not, in fact, eliminate all adverse effects of even the criteria pollutants, 
the NAAQS have become less important. The debate lias shifted to 
questions of technical and economic feasibility, the extent of pollution 
damage compared to the social costs of preventing it, and the like. 
Of course, this deemphasis of the NAAQS in the regulatory process 
was part of the CAA strategy, which incorporated the state imple- 
mentation plan procedure to go from the NAAQS to enforceable 
regulations with as little delay and opportunity for challenge as 

Stationary Sources Controls 

On the whole, the states responded to the requirement for state im- 
plementation plans (SIPs) in a perfunctory manner. For stationary 
sources the regulations are often little more than the old rules on a 
large scale: it is illegal to emit smoke darker than Einjrelmann I.^" 
except for brief periods ; facilities for storing and handling solvents 
must be covered, unless there are good reasons why not; reasonable 
precautions must be taken to prevent dust from blowin,? awav.^* Oc- 
casionally, quantitative limitations are included, such as the emission 
limitations for particulate matter promulgated by EPA in the Ari- 
zona SIP when Arizona failed to do so : according to this standard, 
any process source, handling any dry material of any kind in any 
way, must limit emissions to an amount given by a single mathematical 

In most instances, the SIPs do not even include a good stationary 
source emission inventorv, i.e., a listing of the emission sources in the 

*■• This theory states that there is a levpl of pollution at which effects become sifrnificrint 
more-or-less abruptly. Below the threshold level effects are insignificant ; above the thresh- 
hold they are serious. Thus, the threshold — if there is one — is a natural policy goal what- 
ever the costs of achieving: it. 

^^ An article in the Los Anjrehs TrwM CSunday. Pehruarv 29. 197fi^ entitled "Y.T>\ 
Study — The Findings Got Distorted," accused EPA research aministrators of consciously 
using questionable data and statistical techniques' to demonstrate a threshold where there 
was no real evidence of one. A Congressional hearintr brought forth ^-he expected dis- 
claimers and defenses, and concluded that all was fine. (See Environment Reporter, Current 
Derelopmenfs, vol. 6. No. 51. pp. 2125-6. April 16. 1976). But the EPA researchers are in a 
classic conflict of interest situation, not unlike the nuclear people before the split of the 
AEC into ERDA and NRC. Their bosses are in court defending past agency decisions and 
future plans, and do not like to see research results which question these. It is not sur- 
prising that EPA says its "research in this area is basically directed toward expanding 
the health data base for those pollutants for which NAAQS have been promulgated and 
refining the criteria on which those standards are based," i.e., Is basically defensive. (EPA, 
Progress, 1975, p. 34). 

*' Tlie Rincrelmann index is a measure of the opaqueness of a stream of exhaust gnsses 
usually measured by holding up a standard darkened glass next to the stack and comparing 
darkness; Ringelmann I is twenty percent opacity. Aside from the fact that this test does 
not work too well at night, and that water vapor in the exhaust renders the test worthless. 
It is clearly arbitrary, since a given amount of stuff will have different opacity depending 
on stack size, number of stacks, velocity of exhaust gasses, etc. 

•*s These examples are all from EPA's regulations concerning the SIPs, 40 CFR 51. 
Appendix B gives examples of emission limitations which, in the Administrator's judcrment. 
are attainable. Thus, regulations of the form cited here are not examples of state incom- 
petence ; they reflect EPA's idea of what good emission regulations should look like. 

■'^ In Arizona, the formula states that emissions (in pounds/lir. ) must be less than ?..50 
po.«2 for P<,30 tons/hr., and 17.81 P^-ie for P>30 tons/hr., where P Is process weight of 
dry material In tons/hr. (40 CFR 52.126(b). This is the kind of limitation which Is said 
to be "enforceable" on everything from grain elevators to cement plants to coal distributors. 


state, with estimated emissions of each. In the early plans it was often 
unclear which sources were subject to what regulations, whether or not 
they were in violation, or what they would have to do about it and 

Since the CAA and EPA's regulations clearly require "schedules, 
and timetables for compliance,"' ^^ EPA could not approve SIPs which 
did not include compliance schedules. Eventually some of the states 
produced compliance schedules acceptable to EPA, generally by pub- 
lishing a list of all major sources and saying they would all have to be 
in compliance with all the rules by mid-1975, the date when NAAQS 
had to be met.^^ 'Where a state failed to produce acceptable compliance 
schedules, EPA produced its own, often by a blanket requirement that 
any source subject to regulations must comply by some early date (e.g., 
January 1974) , unless the owner or operator files a schedule promising 
compliance no later than July 1975. In other situations, however, 
EPA has promulgated compliance schedules applicable to broad 
classes of sources and specifying such things as when an abatement 
plan must be prepared, when contracts for fuel or equipment must be 
let, when testing must begin, etc. And all schedules must promise 
compliance by July 1975. 

As a result of this process, for several ^^ars now there have been 
on the books in most states statutes and regulations making certain 
polluting activities unlawful, and specifying in more-or-less detail 
the schedule on which particular sources or classes of sources are sup- 
posed to stop violating those laws by July 1975.^' It is these regulations 
to which EPA referred when it reported that, "With a few notable 
exceptions (e.g., sulfur oxide emission limitations in the State of 
Ohio), all states now [as of 1975] have fully enforceable emission 
limitations atlectiTig stationary sourres.'' ^^ But these "enforceable" 
regulations are typically little more than starting points for negotia- 
tions between the control agency and the indi^ndual sources, each of 
which will argue that: (1) he is in compliance with the regulations; 
(2) if not, it is because the regulation is unreasonable as a general 
rule; (?>) if not, then the regulation is unreasonable in this specific 
case; (4) if not. then it is up to the regulatory agency to tell him how 
he can comply; (5^ if forced to take the steps recommended by the 
regulatory agency, he carmot be held responsible for the results: and 
(6) he needs more time. The regulatory agency, unable to fight every 
battle, will define the regulations so that most sources are in com- 
pliance or can easily comply, and will work out agreements promising 
future action by the worst violators. These agreements then become 
the "enforcoable regulations" and, if they are not complied with, an- 
other round of negotiating is begun. 

^" Seo tostimnnv of Rirhard Avrps. in ''Hearinprs on Implementation of the Tlean Air Act 
Amendments of 1070," Senate Public Works Conim.. Feb. 16, 17, 18. and 23, 1072, at ;^-24. 
Ayres reviewed 2G of the STPs and found them seriously wanting. He blamed the d.-'^icien- 
cies on 0MB weakening of EPA's requirements ; but substantive problems, and not just 
polltioMl Interference in procedures, surely share some of the blame. 

" 1070 OAA. Section 110(a) (2 ) (A ). 

^2 For example. EPA's approval of the Mississippi SIP (40 CFR 52.127) includes a 
24-page list, each page listing 50-60 sources. About half the sources are subject to Mis- 
sissippi regulation APC-S-l in its entirety, the rest to APC-S-1, Section III. The regula- 
tions are effective immediately, and final compliance dates range from April 1973 to July 

•'^•' Or, occaslonallv, as late as 1977, where EPA has given time extensions for attaining 
the primary NAAQS. 

•'•^EPA. rror/rcsK in the J^rerenfion and Control of Air Pollution, 1975, Draft (Herein- 
after referred to as "EPA, Progress"), p. 76. 


This is essentially what is happening in the stationary source air 
pollution control program. There are an estimated 200,000 stationary 
sources subject to SIP emission regulations, about 20,000 of them clas- 
sified as "major", i.e., capable of emitting over 100 tons of pollutant 
per year and, as a class, producing about 85 percent of all air pollu- 
tion from stationary sources. Of these major sources, 14,826 are (and 
probably always have been) in final compliance with SIP regula- 
tions, 1,445 are meeting increments in their compliance schedules, about 
1,200 have yet to be investigated to determine their compliance status, 
and some 2,400 are known to violate SIP emission limitations and 
compliance schedules. EPA estimates that perhaps 130,000 of the 
non-major sources eventually will have to be brought into compliance 
in order to meet I^AAQS.^^ 

Faced with this staggering task and knowing that determined legal 
opposition would smother the agencies and probably destroy many 
of the "enforceable" emission limitations, EPA and the states must 
be cautious and reasonable, and try to work out something with the 
worst emitters. The July 1975 date by which every source was to be 
in compliance has now come and gone, and EPA has begun the en- 
forcement process for the worst polluters, by issuing notices of viola- 
tion, hoping this will shock them into action. If nothing happens within 
30 days,^^ the Administrator "may" issue an order requiring com- 
pliance and, if compliance still is not obtained, may bring a civil ac- 
tion "for appropriate relief." The hope is, of course, that before this 
process gets too far, the state agency or EPA will work out with the 
emitter a new compliance schedule to replace the one which has just 
been violated, and that this time it will be honored. If not, then the 
-case may actually get to court, where EPA has as much to lose as 
does the polluter — if a court throws out the "enforceable emission 
limitation," or rules on some large issue of process or substance, the 
entire program can be disrupted. So EPA and the state try hard to 
settle the issue in a reasonable fashion, short of litigation. 

As of late 1975, EPA had issued about one thousand enforcement 
actions under the CAA, most of them in 1975, since that was when 
final compliance was required in the SIPs." The disposition of all 
those actions has not been determined, but EPA has provided a sum- 
mary of results from enforcement actions against 33 iron and steel 
facilities, most of them initiated in 1974: 32 notices of violation led 
to 18 enforcement orders and four referrals to the Justice Department 
for prosecution. As of late 1975, 2 of the facilities were contending 
they are in final compliance, 18 were on schedules, 7 were negotiating 
schedules, 3 were seeking judicial review of the regulations, and 3 
were "subjects of state/EPA court actions." Thus, after a year or so 
of legal action, the result was about half promises to comply and about 
half delaying actions. 

It is instructive to take a closer look at the compliance schedules, 
since EPA is pointing to these as evidence of progress. A good exam- 
ple is the agreement EPA has extracted from the Jones and Loughlin 

•"' KPA. Projrress, 1075. p. 78. 

5« Rpotion lt?> of thp CAA spoils ont this procodiire for Federal pTiforrpmont. Tf a stnfe 
completely fails to enforce the SIP. EPA can "assume enforcement," in which case the 30- 
dnv notire is not requirerl. Knowing violation of an administrative order is punishable by 
a $2' 000 per dav fine or one year in jail or both. 

57 EPA. Progress, 1975, p. 83. 


Steel Corporation facility in Pittsburgh. EPA issued notices of viola- 
tion to J&L on February 21, 1975, and then spent six months negotiat- 
ing a compliance agreement. This sixty-page document specifies that 
J&L shall : construct basic oxygen furnaces to replace its open hearth 
furnaces, which will be shut down no later than November 1, 1979; 
bring its boilers and coke oven gas emissions into compliance during 
1976 and its ''hot scarfer" into compliance by April 1, 1979; reduce 
smoke emissions from topside pipes, and from coke oven charging 
operations on a detailed scheduKs^^ and 'Vxercise ,<2:oo(l faith and make 
all reasonable efforts" to comply with emission limitations; rehabili- 
tate coke oven stacks on a schedule beginning January 1, 1976, and 
ending January 1, 1980; undertake to develop and test equipment for 
controlling emissions from coke oven stacks and, if it is needed to pre- 
vent violations and it works, to install it within 21 months after a 
violation is established ; design and construct an "enclosed quench car" 
to capture emissions from coke oven pushing by November 1, 1977, 
test it until June 1, 1978, and, if it works, install four more on a de- 
tailed schedule ending December 1, 1980 — if it does not work, a new 
compliance schedule will })e negotiated. There are provisions excusing 
J&L from strict compliance if OSTLA. requirements or "causes be- 
yond the reasonable control" of J&L (i.e., strikes, delays in delivery, 
etc.) intervene. If county standards are relaxed in the future, the 
standards specified in the agreement shall be relaxed; if county stand- 
ards are tightened, the standards specified in the agreement shall not 
be tightened. And, of course, J&L "shall have the right ... to chal- 
lenire the reasonableness or feasibility of any requirement" of the com- 
pliance program, so long as "good faith and reasonable efforts" are 

Generalizing from this EPA "success," it appears that the NAAQS/ 
SIP procedure introduced by the 1970 CAA has not made any funda- 
mental change in the nature of en\4ronmental regulation. The exist- 
ence of health-protecting standards and rigid deadlines for meeting 
them has not produced dramatic progress, and the fact that "enforce- 
able emission limitations" and compliance schedules were on the books 
early has not made compliance automatic or enforcement easy. The 
process remains one of case-by-case negotiation on complex technical 
and economic issues, in which the polluters have the knowledge and 
the resources, and in which the regulators have little choice but to ac- 
cept promises of future actions and good faith. It is really too soon to 
tell whether moving this conventional process to the Federal level is 
enough to make it work at least well enough to get real progress toward 
cleaner air — althoucrh, as discussed presently, the early signs are not 
all that promising. But one should have no illusions about it: for exist- 
ing stationary sources at least, the Clean Air Act is nothing but the 
same old regulatory story, on a massive scale. 

Transportafwn and Land Use Controls 
The SIP process has been, if anyhing, even less successful in deal- 
ing with transportation and land "se controls. The States have taken 

^ For pxample. no more than 5 percent of the pipes shall have visible emissions after 
January 1. 1076; visil»le emissions from eharjrin?: shall not exceed Rinjrf^lmnnn I more 
thiin ;^ niinutos Der lionr after December 1. 1076. and 1 minute per hour after Tune 1. 
1077 ; there shall be visible emissions from no more than 40 percent of coke oven doors 
after September 15. 1076. declining to 34 percent after Februarj 1, 1077 . . . down to 10 
percent after March 1. 1080. 


the position that the Federal emission standards for new automobiles 
would do as much as could be done to control mobile source pollution. 
None of the original SIPs included any extensive use of land-use and 
transportation controls authorized in the Act and, to the extent they 
did promise these measures, contained inadequate legal authority to 
implement them. Kecognizing the magnitude of the problem, EPA 
Administrator Euckelshaus tried to be reasonable and give the States 
an additional two years to prepare transportation control plans 
(TCPs) and to obtain legal authority to implement them. But two 
court cases,^® brought by citizens groups under the citizen suit provi- 
sions of the CAA, forced EPA to rescind this extension of time. Thus, 
EPA had no choice but to require the States to submit SIPs, including 
t]*ansportation controls where necessary, which would accomplish the 
NAAQS by 1975, and to produce its own plans where the State failed 
to do so. This led to the absurd situation in which Ruckelshaus was 
forced to promulgate a SIP for California which required reduction 
of gasoline usage in the Los Angeles basin by 87 percent during the 
summer smog season.®^ 

EPA has tried to get the states to propose and implement measures 
to control transportation as part of an effort to accomplish NAAQS, 
and has failed. AYhen EPA tried to impose its own plans, which re- 
quired surcharges on parking, the management ®^ of parking supply, 
and the requirement that preferential bus and carpool lanes be pro- 
vided on highways. Congress and the courts intervened : the ESECA 
amendments make it difficult or illegal for EPA to impose or require 
such measures ; ^^ EPA's appropriation bill for fiscal year 1975 delayed 
the agency's authority to require parking management plans ; ^^ and 
the constitutionality of EPA's attempts to force a state to regulate 
its citizens is being challenged in the courts.^* Most recently, the 1977 
Amendments to the CAA have recognized the fact that the NAAQS 
are not going to be met everywhere, and allow the Administrator to 
extend the compliance date for NAAQS by five years anywhere, and 
by as much as ten years in regions where transportation control meas- 
ures are necessary for attainment. 

The complex or indirect source controls have suffered a similar fate. 
After issuing final regulations in February 1974, EPA delayed fed- 
eral enforcement several times. Then ESECA, the 1975 appropria- 
tions bill for EPA, and finally the 1977 CAA Amendments restricted 
EPA's ability to enforce indirect controls. 

National Emission Standards for Stationary Sources 

New source performance standards (NSPS), applicable to any new 
or "modified" source, are to be set and enforced by EPA, on the basis 
of the best system of control available, for any category of emitters 
which the Administrator determines to be a significant contributor to 

""■' City of Riverside v. Euckelshaus, 4 EEC 1728 (CD. Calif. 1972) and NRDG v. EPA, 
4 ERC 1945 (D.C. Cir. 1973). 

^" See discussions in E. Chernow, "Implementinfr the Clean Air Act in Los Angeles • The 
Duty to Achieve the Impossible," Ecology Law Quarterly, Vol. 4, No. 3 (1975) and in 
P. B. Downing and G. Brady, "Implementing the Clean Air Act : A Case Study of Oxidant 
Control in Los Angeles," Virginia Polytechnic Institute (July, 1976), Draft. 

" Defined to mean the control of new parking facilities, usually via a permit system in 
which pf^rmits would he issued on the basis of regional air quality considerations.' 

" Section 110(c) (2) of the CAA, as amended bv ESECA. 

63 Agriculture. Environmental and Consumer Protection Appropriation Act, PL 93-563. 
Sec. 510 (1975). 

^* See discussion in EPA, Progress, 1975, p. 101. 


air pollution. Xational emission standards for hazardous air pol- 
lutants (XESHAPS) are to be established for any pollutant which 
the Administrator decides to put on a list of hazardous pollutants, and 
must be adequate to protect public health without regard to cost of 
compliance; a source cleans up or shuts down. These provisions, on 
their face, would seem to provide the basis for an effective and rela- 
tively simple regulatory program run directly by EPA, without all the 
complications of the SIPs. 

New Source Performance Standards 

The theory beliind the NSPS provision is simple enough — EPA 
will pick a standard for a category or subcategory, using only tech- 
nological and economic criteria, establish its enforceability, and every 
new source will meet it or not operate. Unfortunately, even for new 
sources things are not so simple, given the complexity of industrial 
operations and tJie diversity among sources. Thus, the five NSPS 
promulgated by EPA in 1971 were still being litigated in 1976. 

The issues involved in some of the early XSPS challenges are in- 
structive. For fossil fuel fired steam generators, i.e., power plants, the 
XSPS for sulfur dioxide was set at 0.8 Ib./^NOIBtu for liquid fuel and 
1.2 lb./]\OIBtu for solid fuel, in an attempt to force the use of low- 
sulfur coal or the adoption of stack-gas scrubbers on new coal-fired 
boilers; "'' the XSPS wore remanded back to EPA for reconsideration, 
Avliere they have been debated for several years. In the case of the port- 
land cement XSPS, the court ruled that EPA could properly impose a 
stricter standard on one industry than on another where technology 
was unavailable, thereby penalizing the more progressive industry; 
the court also noted that in competing industries, EPA would have 
to take competitive aspects into account. ^^ The same case also chal- 
lenged the Ringelmann standards as arbitrary, and challenged the 
application of the XSPS during periods of start-up, shut-down and 
malfunction; on remand, EPA decided the Ringelm.ann standard is 
viable, but changed its XSPS cei-tification procedures so that tests 
will be run only during ''normal" operation, i.e., when the plant op- 
erator has everything finely tuned. 

Thus, the familiar regulatory pattern has developed, even in the 
simple case of setting technology-based standards for new sources. 
EPA must consider a broad range of technical and economic issues, 
make many distinctions, make and defend many difficult judgments. 
One important difference here is that delay and indecision in the proc- 
ess prevents rather than allows pollution from a specific source, be- 
cause an EPA permit is require to begin construction. Judged by its 
ability to prevent pollution from neio sources, the XSPS provision 
may be successful. But judged by its ability to manage the expansion 
of industry and the replacement of old sources by new with a minimum 

^■■' 40 CFR 60.4. The emission fijrures refer to pounds of SO2 emitted per million Btus of 
fuel burned. For coal, 1.2 lbs. SO^/MBtu is equivalent to coal containing less than 1 per- 
cent sulfur, as little as 0.5 percent sulfur for lower Btu western coal. Why do the XSPS 
allow luilf a.irain as much SO2 per heat unit for coal as for oil? Because coal is inherently 
dirtier. Should it work out that coal-firing is marginally cheaper than oil-firing, cost- 
niiuiniizing ent«'rprises would choose coal over oil. giving SOo emissions 50 percent higher 
than Thev need be with no economic advantage to society. 

'^n'ortlaud Cement v. Kuckel.sJiaiis. 5 P]RC 1593 (D.C. Circuit 7AE9). See discussion in 
Arthur Ferguson. '"Direct Federal Controls," in Ecology Law Quarterly, Vol. 4, Xo. 3, 
1975, pp. 649-650. 


of unnecessary delay, uncertainty, conflict, and government interven- 
tion, the NSPS provision leaves something to be desired. 

National Emission Standards for Hazardous Air Pollutants 
Implementation of the provision concerning national emission stand- 
ards for hazardous air pollutants (NESHAP) began in March 1971, 
when, as required by the CAA, the Administrator published a list 
of substances for which he proposed to set a NESHAP. There were 
three substances on the list — asbestos, beryllium and mercury. When 
EPA was a year late in promulgating standards for the substances 
on the list, a citizens' suit was brought and, in 1973, the three 
NESHAPS were promulgated. In 1975 some revisions in these stand- 
ards were promulgated, and a fourth substance (vinyl chloride) added 
to the list. By 1975, 620 sources were subject to an appro v^ed 

A principal difficulty with the NESHAPs has been deciding how 
one relates a national emission standard to a harmful concentration 
of pollutant in the air. For the four pollutants on EPA's current list 
(and, indeed, for most "toxic" substances and probably for all pollut- 
ants) there is no demonstrated "safe" level. The most one can say is 
that it gets harder to prove adverse effects at lower concentration 
levels, where measurement techniques, experimental methods, and 
statistical procedures are less well-developed. And even if one could 
find an ambient concentration level which was safe, how does one set 
a national emission standard (i.e., a standard limiting the concentra- 
tion or flow rate of the substance in the exhaust stream), applicable to 
sources of any size and technology wliich guarantees the ambient level 
does not exceed the safe level ? 

These logical and informational difficulties go far to explain why the 
NESHAP provision has been used so little. For the original three 
substances on the list, EPA has, in effect, set local ambient air stand- 
ards and required each source in various classes to demonstrate that 
its emissions are not causing concentrations above those levels. For 
asbestos, the NESHAP specifies such things as how asbestos-insulated 
buildings are to be demolished. The proposed vinyl-chloride NESHAP 
is based more on a "best available technology" philosophy than on a 
"protect health at any cost" philosophy — perhaps a sensible move, 
but one more easily done in other ways.^^ Thus, the NESHAP pro- 
vision has not proved very useful, and was largely supplanted in 1977 
by new toxic substances legislation. 

Uniform Percentage Reduction Standards 
The 1977 Amendments to the CAA introduced a new element in the 
Federal air pollution control program, by requiring the use of tech- 
nological controls on all new sources burning solid fossil fuels, inde- 
pendent of ambient air quality or of the uncontrolled emission levels. 
For sulfur, nitrogen oxides and particulates, EPA is to specify the 
fixed percentages by which emissions must be reduced below the level 
they would be in the absence of any technological control. Thus, the 

^'^ EPA, Progress, 1975, p. 86. See discussion in A. B. Ferguson. Jr., "Direct Federal 
Controls : NSPS and Hazardous Emissions," Ecology Law Quarterly, Vol. 4, No. 3 (1975). 

•''^ It would seem more sensible and lojral to use the NSPS to implempnt a "hest tech- 
nology everywhere" goal for a specific pollutant. Section 111 of the CAA allows EPA to 
set a NSPS for any pollutant it chooses, and then to require existing sources to meet 
similar emission standards for those pollutants regulated by a NSPS but not by a NAAQS. 
Better yet would be an emissions price applicable to all emissions everywhere. 


use of inherently clean fuels will no longer be an acceptable method 
of emission control for solid fossil fuel combustion, and the relevance 
of the ambient air quality standards has been further reduced. 

How EPA will implement this Congressional direction remains to 
be seen. For nitrogen oxides and particulates, which are created in 
the coal handling/combustion process itself, there is no logical way 
even to define such standards. 

For sulfur, which is inherent in the coal, the concept of a uniform 
percentage reduction can be defined, but becomes extremely difficult to 
implement, especially in situations where coal from various sources is 
blended, processed, converted into other forms and used in various 
combustion processes, with emissions at each stage; the difficulty of 
defining and demonstrating compliance with the standards will itself 
become a constraint on adoption of such systems. And some promising 
coal-using technologies, such as fluidized bed combustion and solvent 
refining, may never be developed if the sulfur-removal percentage is set 
too strictly; the risk is too great that they will control only (say) 89 
percent of emissions instead of the required level of (say) 90 percent. 
Thus, large costs will be imposed for little or no environmental gain. 
It has been suggested that the real motive behind this provision was to 
enhance the competitive position of higher-sulfur eastern coal relative 
to western coal, for political reasons. Even to accomplish this ques- 
tionable goal, the uniform-percentage-reduction standard is a crude 
device, the only advantage of which is a political one : it obscures the 
real objective. 

This provision is yet another example of regulatory logic (or 
il logic) at work. Unhappy with events. Congress has tried to cut 
through the complex processes of the CAA (which were themselves 
initially regarded as a major simplification), with a simple rule which 
makes it unnecessary to make distinctions. However, it probably will 
soon be clear that the costs of ignoring economic and technical com- 
plexities are too great, and the regulatory /legal process will bog down, 
forcing Congress to reconsider. The result will be several years of 
delay and uncertainty, followed by adoption of a complex regulatory 
procedure which tries to accomplish the "best technology everywhere" 
goal while making the appropriate technological and economic dis- 
tinctions, and which (unless regulatory processes are supplemented 
with market forces) does so poorly. 

Nondetemoration of Air Quality 

When the Supreme Court ruled in 1973 that EPA must prevent 
''significant dc'terioration'' of air quality .'^'■^ two things were apparent: 
it would be difficult to turn this general requirement into an opera- 
tional program ; and any program which was "reaP' would have sig- 
nificant social and economic effects. A nationwide limit on increasing 
emissions in clean air areas — about 80 percent of the land area of the 
U.S. — was thought to be intolerable. EPA proposed a number of ap- 
proaches to the problem for discussion, and delayed doing anything 
until threatened by a court order. 

Finally, EPA proposed a zoning plan, in which the clean air re- 
gions of the country would be divided into three classes, depending 
on how much air pollution would be allowed to increase in the region.^** 

bi» See tpxt and footnote, p. 279 above. 

^ This discussion borrows from Disselhast, Ecology Law Quarterly, 1975. 

8.1-944 — 78 20 


Initially, all clean air regions would be in Class II, where increases 
in emissions would be allowed if new sources used best available 
control technology. State Governors could then classify regions